DYNARE

User Guide
Tommaso Mancini Griffoli, 2007-2008
An introduction to
the solution & estimation of DSGE models
Dynare v4 - User Guide
Public beta version
Tommaso Mancini Griffoli
tommaso.mancini@stanfordalumni.org
This draft: June 2010
iii
Copyright c 2007-2010 Tommaso Mancini Griffoli
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3 or any
later version published by the Free Software Foundation; with no Invariant
Sections, no Front-Cover Texts, and no Back-Cover Texts.
A copy of the license can be found at: http://www.gnu.org/licenses/
fdl.txt
Contents
Contents iv
List of Figures vii
1 Introduction 1
1.1 About this Guide - approach and structure . . . . . . . . . . . 1
1.2 What is Dynare? . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Additional sources of help . . . . . . . . . . . . . . . . . . . . . 4
1.4 Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.5 v4, what’s new and backward compatibility . . . . . . . . . . . 5
2 Installing Dynare 7
2.1 Dynare versions . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 System requirements . . . . . . . . . . . . . . . . . . . . . . . . 7
2.3 Installing Dynare . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.4 MATLAB particularities . . . . . . . . . . . . . . . . . . . . . . 8
3 Solving DSGE models - basics 9
3.1 A fundamental distinction . . . . . . . . . . . . . . . . . . . . . 9
3.1.1 NOTE! Deterministic vs stochastic models . . . . . . . . 10
3.2 Introducing an example . . . . . . . . . . . . . . . . . . . . . . 11
3.3 Dynare .mod file structure . . . . . . . . . . . . . . . . . . . . . 15
3.4 Filling out the preamble . . . . . . . . . . . . . . . . . . . . . . 15
3.4.1 The deterministic case . . . . . . . . . . . . . . . . . . . 16
3.4.2 The stochastic case . . . . . . . . . . . . . . . . . . . . . 16
3.4.3 Comments on your first lines of Dynare code . . . . . . 17
3.5 Specifying the model . . . . . . . . . . . . . . . . . . . . . . . . 18
3.5.1 Model in Dynare notation . . . . . . . . . . . . . . . . . 18
3.5.2 General conventions . . . . . . . . . . . . . . . . . . . . 19
3.5.3 Notational conventions . . . . . . . . . . . . . . . . . . . 19
3.5.4 Timing conventions . . . . . . . . . . . . . . . . . . . . 19
3.5.5 Conventions specifying non-predetermined variables . . 20
3.5.6 Linear and log-linearized models . . . . . . . . . . . . . 20
3.6 Specifying steady states and/or initial values . . . . . . . . . . 21
iv
CONTENTS v
3.6.1 Stochastic models and steady states . . . . . . . . . . . 21
3.6.2 Deterministic models and initial values . . . . . . . . . . 23
3.6.3 Finding a steady state . . . . . . . . . . . . . . . . . . . 23
3.6.4 Checking system stability . . . . . . . . . . . . . . . . . 24
3.7 Adding shocks . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.7.1 Deterministic models - temporary shocks . . . . . . . . 25
3.7.2 Deterministic models - permanent shocks . . . . . . . . 25
3.7.3 Stochastic models . . . . . . . . . . . . . . . . . . . . . 27
3.8 Selecting a computation . . . . . . . . . . . . . . . . . . . . . . 27
3.8.1 For deterministic models . . . . . . . . . . . . . . . . . . 28
3.8.2 For stochastic models . . . . . . . . . . . . . . . . . . . 28
3.9 The complete .mod file . . . . . . . . . . . . . . . . . . . . . . . 31
3.9.1 The stochastic model . . . . . . . . . . . . . . . . . . . . 31
3.9.2 The deterministic model (case of temporary shock) . . . 32
3.10 File execution and results . . . . . . . . . . . . . . . . . . . . . 33
3.10.1 Results - stochastic models . . . . . . . . . . . . . . . . 33
3.10.2 Results - deterministic models . . . . . . . . . . . . . . 34
4 Solving DSGE models - advanced topics 37
4.1 Dynare features and functionality . . . . . . . . . . . . . . . . . 37
4.1.1 Other examples . . . . . . . . . . . . . . . . . . . . . . . 37
4.1.2 Alternative, complete example . . . . . . . . . . . . . . 38
4.1.3 Finding, saving and viewing your output . . . . . . . . . 41
4.1.4 Referring to external files . . . . . . . . . . . . . . . . . 42
4.1.5 Infinite eigenvalues . . . . . . . . . . . . . . . . . . . . . 43
4.2 Files created by Dynare . . . . . . . . . . . . . . . . . . . . . . 43
4.3 Modeling tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.3.1 Stationarizing your model . . . . . . . . . . . . . . . . . 44
4.3.2 Expectations taken in the past . . . . . . . . . . . . . . 44
4.3.3 Infinite sums . . . . . . . . . . . . . . . . . . . . . . . . 44
4.3.4 Infinite sums with changing timing of expectations . . . 46
5 Estimating DSGE models - basics 47
5.1 Introducing an example . . . . . . . . . . . . . . . . . . . . . . 47
5.2 Declaring variables and parameters . . . . . . . . . . . . . . . . 48
5.3 Declaring the model . . . . . . . . . . . . . . . . . . . . . . . . 48
5.4 Declaring observable variables . . . . . . . . . . . . . . . . . . . 49
5.5 Specifying the steady state . . . . . . . . . . . . . . . . . . . . 49
5.6 Declaring priors . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.7 Launching the estimation . . . . . . . . . . . . . . . . . . . . . 52
5.8 The complete .mod file . . . . . . . . . . . . . . . . . . . . . . . 55
5.9 Interpreting output . . . . . . . . . . . . . . . . . . . . . . . . . 57
5.9.1 Tabular results . . . . . . . . . . . . . . . . . . . . . . . 57
5.9.2 Graphical results . . . . . . . . . . . . . . . . . . . . . . 57
6 Estimating DSGE models - advanced topics 61
6.1 Alternative and non-stationary example . . . . . . . . . . . . . 61
6.1.1 Introducing the example . . . . . . . . . . . . . . . . . . 61
6.1.2 Declaring variables and parameters . . . . . . . . . . . . 66
6.1.3 The origin of non-stationarity . . . . . . . . . . . . . . . 66
6.1.4 Stationarizing variables . . . . . . . . . . . . . . . . . . 67
6.1.5 Linking stationary variables to the data . . . . . . . . . 68
6.1.6 The resulting model block of the .mod file . . . . . . . . 68
6.1.7 Declaring observable variables . . . . . . . . . . . . . . . 69
6.1.8 Declaring trends in observable variables . . . . . . . . . 69
6.1.9 Declaring unit roots in observable variables . . . . . . . 70
6.1.10 Specifying the steady state . . . . . . . . . . . . . . . . 71
6.1.11 Declaring priors . . . . . . . . . . . . . . . . . . . . . . 71
6.1.12 Launching the estimation . . . . . . . . . . . . . . . . . 71
6.1.13 The complete .mod file . . . . . . . . . . . . . . . . . . . 72
6.1.14 Summing it up . . . . . . . . . . . . . . . . . . . . . . . 74
6.2 Comparing models based on their posterior distributions . . . . 74
6.3 Where is your output stored? . . . . . . . . . . . . . . . . . . . 75
7 Solving DSGE models - Behind the scenes of Dynare 77
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
7.2 What is the advantage of a second order approximation? . . . . 77
7.3 How does dynare solve stochastic DSGE models? . . . . . . . . 78
8 Estimating DSGE models - Behind the scenes of Dynare 81
8.1 Advantages of Bayesian estimation . . . . . . . . . . . . . . . . 81
8.2 The basic mechanics of Bayesian estimation . . . . . . . . . . . 82
8.2.1 Bayesian estimation: somewhere between calibration and
maximum likelihood estimation - an example . . . . . . 84
8.3 DSGE models and Bayesian estimation . . . . . . . . . . . . . . 85
8.3.1 Rewriting the solution to the DSGE model . . . . . . . 85
8.3.2 Estimating the likelihood function of the DSGE model . 86
8.3.3 Finding the mode of the posterior distribution . . . . . 87
8.3.4 Estimating the posterior distribution . . . . . . . . . . . 87
8.4 Comparing models using posterior distributions . . . . . . . . . 90
9 Optimal policy under commitment 93
10 Troubleshooting 95
vi
List of Figures vii
Bibliography 97
List of Figures
1.1 Dynare, a bird’s eyeview . . . . . . . . . . . . . . . . . . . . . . . . 3
3.1 Structure of the .mod file . . . . . . . . . . . . . . . . . . . . . . . 16
6.1 CIA model illustration . . . . . . . . . . . . . . . . . . . . . . . . . 62
6.2 Steps of model estimation . . . . . . . . . . . . . . . . . . . . . . . 74
8.1 Illustration of the Metropolis-Hastings algorithm . . . . . . . . . . 89
Work in Progress!
This is the second version of the Dynare User Guide which is still work in
progress! This means two things. First, please read this with a critical eye
and send me comments! Are some areas unclear? Is anything plain wrong?
Are some sections too wordy, are there enough examples, are these clear? On
the contrary, are there certain parts that just click particularly well? How can
others be improved? I’m very interested to get your feedback.
The second thing that a work in progress manuscript comes with is a few
internal notes. These are mostly placeholders for future work, notes to myself
or others of the Dynare development team, or at times notes to you - our read-
ers - to highlight a feature not yet fully stable. Any such notes are marked
with two stars (**).
Thanks very much for your patience and good ideas. Please write either
direclty to myself: tommaso.mancini@stanfordalumni.org, or preferably on
the Dynare Documentation Forum available in the Dynare Forums.
ix
Contacts and Credits
Dynare was originally developed by Michel Juillard in Paris, France. Cur-
rently, the development team of Dynare is composed of
• St´ephane Adjemian (stephane.adjemian“AT”ens.fr)
• Houtan Bastani (houtan.bastani“AT”ens.fr)
• Michel Juillard (michel.juillard“AT”mjui.fr)
• Ferhat Mihoubi (ferhat.mihoubi“AT”univ-evry.fr)
• George Perendia (george“AT”perendia.orangehome.co.uk)
• Marco Ratto (marco.ratto“AT”jrc.ec.europa.eu)
• S´ebastien Villemot (sebastien.villemot“AT”ens.fr)
Several parts of Dynare use or have strongly benefited from publicly avail-
able programs by G. Anderson, F. Collard, L. Ingber, O. Kamenik, P. Klein,
S. Sakata, F. Schorfheide, C. Sims, P. Soederlind and R. Wouters.
Finally, the development of Dynare could not have come such a long ways
withough an active community of users who continually pose questions, re-
port bugs and suggest new features. The help of this community is gratefully
acknowledged.
The email addresses above are provided in case you wish to contact any
one of the authors of Dynare directly. We nonetheless encourage you to first
use the Dynare forums to ask your questions so that other users can benefit
from them as well; remember, almost no question is specific enough to interest
just one person, and yours is not the exception!
xi
Chapter 1
Introduction
Welcome to Dynare!
1.1 About this Guide - approach and structure
This User Guide aims to help you master Dynare’s main functionalities,
from getting started to implementing advanced features. To do so, this Guide
is structured around examples and offers practical advice. To root this un-
derstanding more deeply, though, this Guide also gives some background on
Dynare’s algorithms, methodologies and underlying theory. Thus, a secondary
function of this Guide is to serve as a basic primer on DSGE model solving
and Bayesian estimation.
This Guide will focus on the most common or useful features of the pro-
gram, thus emphasizing depth over breadth. The idea is to get you to
use 90% of the program well and then tell you where else to look if you’re
interested in fine tuning or advanced customization.
This Guide is written mainly for an advanced economist - like a pro-
fessor, graduate student or central banker - needing a powerful and flexible
program to support and facilitate his or her research activities in a variety
of fields. The sophisticated computer programmer, on the one hand, or the
specialist of computational economics, on the other, may not find this Guide
sufficiently detailed.
We recognize that the “advanced economist” may be either a beginning
or intermediate user of Dynare. This Guide is written to accommodate both.
If you’re new to Dynare, we recommend starting with chapters 3 and 5,
which introduce the program’s basic features to solve (including running im-
pulse response functions) and estimate DSGE models, respectively. To do
1
2 CHAPTER 1. INTRODUCTION
so, these chapters lead you through a complete hands-on example, which we
recommend following from A to Z, in order to “learn by doing”. Once you
have read these two chapters, you will know the crux of Dynare’s functionality
and (hopefully!) feel comfortable using Dynare for your own work. At that
point, though, you will probably find yourself coming back to the User Guide
to skim over some of the content in the advanced chapters to iron out details
and potential complications you may run into.
If you’re instead an intermediate user of Dynare, you will most likely
find the advanced chapters, 4 and 6, more appropriate. These chapters cover
more advanced features of Dynare and more complicated usage scenarios. The
presumption is that you would skip around these chapters to focus on the top-
ics most applicable to your needs and curiosity. Examples are therefore more
concise and specific to each feature; these chapters read a bit more like a ref-
erence manual.
We also recognize that you probably have had repeated if not active ex-
posure to programming and are likely to have a strong economic background.
Thus, a black box solution to your needs is inadequate. To hopefully address
this issue, the User Guide goes into some depth in covering the theoreti-
cal underpinnings and methodologies that Dynare follows to solve
and estimate DSGE models. These are available in the “behind the scenes of
Dynare” chapters 7 and 8. These chapters can also serve as a basic primer
if you are new to the practice of DSGE model solving and Bayesian estimation.
Finally, besides breaking up content into short chapters, we’ve introduced
two different markers throughout the Guide to help streamline your reading.
• TIP! introduces advice to help you work more efficiently with Dynare
or solve common problems.
• NOTE! is used to draw your attention to particularly important infor-
mation you should keep in mind when using Dynare.
1.2 What is Dynare?
Before we dive into the thick of the “trees”, let’s have a look at the “forest”
from the top . . . just what is Dynare?
Dynare is a powerful and highly customizable engine, with an
intuitive front-end interface, to solve, simulate and estimate DSGE
models.
1.2. WHAT IS DYNARE? 3
Figure 1.1: The .mod file being read by the Dynare pre-processor, which then
calls the relevant Matlab routines to carry out the desired operations and
display the results.
In slightly less flowery words, it is a pre-processor and a collection of Mat-
lab routines that has the great advantages of reading DSGE model equations
written almost as in an academic paper. This not only facilitates the inputting
of a model, but also enables you to easily share your code as it is straightfor-
ward to read by anyone.
Figure 1.2 gives you an overview of the way Dynare works. Basically, the
model and its related attributes, like a shock structure for instance, is writ-
ten equation by equation in an editor of your choice. The resulting file will
be called the .mod file. That file is then called from Matlab. This initiates
the Dynare pre-processor which translates the .mod file into a suitable input
for the Matlab routines (more precisely, it creates intermediary Matlab or C
files which are then used by Matlab code) used to either solve or estimate the
model. Finally, results are presented in Matlab. Some more details on the
internal files generated by Dynare is given in section 4.2 in chapter 4.
Each of these steps will become clear as you read through the User Guide,
but for now it may be helpful to summarize what Dynare is able to do:
4 CHAPTER 1. INTRODUCTION
• compute the steady state of a model
• compute the solution of deterministic models
• compute the first and second order approximation to solutions of stochas-
tic models
• estimate parameters of DSGE models using either a maximum likelihood
or a Bayesian approach
• compute optimal policies in linear-quadratic models
1.3 Additional sources of help
While this User Guide tries to be as complete and thorough as possible, you
will certainly want to browse other material for help, as you learn about new
features, struggle with adapting examples to your own work, and yearn to ask
that one question whose answer seems to exist no-where. At your disposal,
you have the following additional sources of help:
• Reference Manual: this manual covers all Dynare commands, giving
a clear definition and explanation of usage for each. The User Guide
will often introduce you to a command in a rather loose manner (mainly
through examples); so reading corresponding command descriptions in
the Reference Manual is a good idea to cover all relevant details.
• Official online examples: the Dynare website includes other examples
- usually well documented - of .mod files covering models and method-
ologies introduced in recent papers.
• Dynare forums: this lively online discussion forum allows you to ask
your questions openly and read threads from others who might have run
into similar difficulties.
• Frequently Asked Questions (FAQ): this section of the Dynare site
emphasizes a few of the most popular questions in the forums.
• DSGE.net: this website, run my members of the Dynare team, is a
resource for all scholars working in the field of DSGE modeling. Besides
allowing you to stay up to date with the most recent papers and possi-
bly make new contacts, it conveniently lists conferences, workshops and
seminars that may be of interest.
1.4. NOMENCLATURE 5
1.4 Nomenclature
To end this introduction and avoid confusion in what follows, it is worthwhile
to agree on a few definitions of terms. Many of these are shared with the
Reference Manual.
• Integer indicates an integer number.
• Double indicates a double precision number. The following syntaxes
are valid: 1.1e3, 1.1E3, 1.1E-3, 1.1d3, 1.1D3
• Expression indicates a mathematical expression valid in the underlying
language (e.g. Matlab).
• Variable name indicates a variable name. NOTE! These must start
with an alphabetical character and can only contain other alphabetical
characters and digits, as well as underscores ( ). All other characters,
including accents, and spaces, are forbidden.
• Parameter name indicates a parameter name which must follow the
same naming conventions as above.
• Filename indicates a file name valid in your operating system. Note
that Matlab requires that names of files or functions start with alpha-
betical characters; this concerns your Dynare .mod files.
• Command is an instruction to Dynare or other program when specified.
• Options or optional arguments for a command are listed in square
brackets [ ] unless otherwise noted. If, for instance, the option must
be specified in parenthesis in Dynare, it will show up in the Guide as
[(option)].
• Typewritten text indicates text as it should appear in Dynare code.
1.5 v4, what’s new and backward compatibility
The current version of Dynare - for which this guide is written - is version
4. With respect to version 3, this new version introduces several important
features, as well as improvements, optimizations of routines and bug fixes.
The major new features are the following:
• Analytical derivatives are now used everywhere (for instance, in the
Newton algorithm for deterministic models and in the linearizations nec-
essary to solve stochastic models). This increases computational speed
significantly. The drawback is that Dynare can now handle only a lim-
ited set of functions, although in nearly all economic applications this
should not be a constraint.
6 CHAPTER 1. INTRODUCTION
• Variables and parameters are now kept in the order in which they are
declared whenever displayed and when used internally by Dynare. Recall
that in version 3, variables and parameters where at times in their order
of declaration and at times in alphabetical order. NOTE! This may
cause some problems of backward compatibility if you wrote programs
to run off Dynare v3 output.
• The names of many internal variables and the organization of output
variables has changed. These are enumerated in details in the relevant
chapters. The names of the files internally generated by Dynare have
also changed. (** more on this when explaining internal file structure -
TBD)
• The syntax for the external steady state file has changed. This is cov-
ered in more details in chapter 3, in section 3.6.3. NOTE! You will
unfortunately have to slightly amend any old steady state files you may
have written.
• Speed. Several large-scale improvements have been implemented to
speed up Dynare. This should be most noticeable when solving de-
terministic models, but also apparent in other functionality.
Chapter 2
Installing Dynare
2.1 Dynare versions
The current version of Dynare (4.1) runs on both MATLAB and GNU
Octave.
There used to be versions of Dynare for Scilab and Gauss. Development
of the Scilab version stopped after Dynare version 3.02 and that for Gauss
after Dynare version 1.2.
This User Guide will exclusively focus on Dynare version 4.0 and
later.
You may also be interested by another program, Dynare++, which is
a standalone C++ program specialized in computing k-order approximations
of dynamic stochastic general equilibrium models. Note that Dynare++ is
distributed along with Dynare since version 4.1. See the Dynare++ webpage
for more information.
2.2 System requirements
Dynare can run on Microsoft Windows, as well as Unix-like operating systems,
in particular GNU/Linux and Mac OS X. If you have questions about the
support of a particular platform, please ask your question on Dynare forums.
To run Dynare, it is recommended to allocate at least 256MB of RAM
to the platform running Dynare, although 512MB is preferred. Depending on
the type of computations required, like the very processor intensive Metropolis
Hastings algorithm, you may need up to 1GB of RAM to obtain acceptable
computational times.
2.3 Installing Dynare
Please refer to the section entitled “Installation and configuration” in the
Dynare reference manual.
7
8 CHAPTER 2. INSTALLING DYNARE
2.4 MATLAB particularities
A question often comes up: what special MATLAB toolboxes are necessary
to run Dynare? In fact, no additional toolbox is necessary for running most of
Dynare, except maybe for optimal simple rules (see chapter 9), but even then
remedies exist (see the Dynare forums for discussions on this, or to ask your
particular question). But if you do have the ‘optimization toolbox’ installed,
you will have additional options for solving for the steady state (solve algo
option) and for searching for the posterior mode (mode compute option), both
of which are defined later.
Chapter 3
Solving DSGE models - basics
This chapter covers everything that leads to, and stems from, the solution
of DSGE models; a vast terrain. That is to say that the term “solution”
in the title of the chapter is used rather broadly. You may be interested in
simply finding the solution functions to a set of first order conditions stemming
from your model, but you may also want to go a bit further. Typically, you
may be interested in how this system behaves in response to shocks, whether
temporary or permanent. Likewise, you may want to explore how the system
comes back to its steady state or moves to a new one. This chapter covers all
these topics. But instead of skipping to the topic closest to your needs, we
recommend that you read this chapter chronologically, to learn basic Dynare
commands and the process of writing a proper .mod file - this will serve as a
base to carry out any of the above computations.
3.1 A fundamental distinction
Before speaking of Dynare, it is important to recognize a distinction in model
types. This distinction will appear throughout the chapter; in fact, it is so
fundamental, that we considered writing separate chapters altogether. But
the amount of common material - Dynare commands and syntax - is notable
and writing two chapters would have been overly repetitive. Enough suspense;
here is the important question: is your model stochastic or determinis-
tic?
The distinction hinges on whether future shocks are known. In de-
terministic models, the occurrence of all future shocks is known exactly at
the time of computing the model’s solution. In stochastic models, instead,
only the distribution of future shocks is known. Let’s consider a shock to a
model’s innovation only in period 1. In a deterministic context, agents will
take their decisions knowing that future values of the innovations will be zero
in all periods to come. In a stochastic context, agents will take their decisions
9
10 CHAPTER 3. SOLVING DSGE MODELS - BASICS
knowing that the future value of innovations are random but will have zero
mean. This isn’t the same thing because of Jensen’s inequality. Of course, if
you consider only a first order linear approximation of the stochastic model,
or a linear model, the two cases become practically the same, due to certainty
equivalence. A second order approximation will instead lead to very different
results, as the variance of shocks will matter.
The solution method for each of these model types differs significantly. In
deterministic models, a highly accurate solution can be found by numerical
methods. The solution is nothing more than a series of numbers that match
a given set of equations. Intuitively, if an agent has perfect foresight, she can
specify today - at the time of making her decision - what each of her precise
actions will be in the future. In a stochastic environment, instead, the best
the agent can do is specify a decision, policy or feedback rule for the future:
what will her optimal actions be contingent on each possible realization of
shocks. In this case, we therefore search for a function satisfying the model’s
first order conditions. To complicate things, this function may be non-linear
and thus needs to be approximated. In control theory, solutions to determin-
istic models are usually called “closed loop” solutions, and those to stochastic
models are referred to as “open loop”.
Because this distinction will resurface again and again throughout the
chapter, but also because it has been a source of significant confusion in the
past, the following gives some additional details.
3.1.1 NOTE! Deterministic vs stochastic models
Deterministic models have the following characteristics:
1. As the DSGE (read, “stochastic”, i.e. not deterministic!) literature
has gained attention in economics, deterministic models have become
somewhat rare. Examples include OLG models without aggregate un-
certainty.
2. These models are usually introduced to study the impact of a change in
regime, as in the introduction of a new tax, for instance.
3. Models assume full information, perfect foresight and no uncertainty
around shocks.
4. Shocks can hit the economy today or at any time in the future, in which
case they would be expected with perfect foresight. They can also last
one or several periods.
5. Most often, though, models introduce a positive shock today and zero
shocks thereafter (with certainty).
3.2. INTRODUCING AN EXAMPLE 11
6. The solution does not require linearization, in fact, it doesn’t even really
need a steady state. Instead, it involves numerical simulation to find the
exact paths of endogenous variables that meet the model’s first order
conditions and shock structure.
7. This solution method can therefore be useful when the economy is far
away from steady state (when linearization offers a poor approximation).
Stochastic models, instead, have the following characteristics:
1. These types of models tend to be more popular in the literature. Exam-
ples include most RBC models, or new keynesian monetary models.
2. In these models, shocks hit today (with a surprise), but thereafter their
expected value is zero. Expected future shocks, or permanent changes
in the exogenous variables cannot be handled due to the use of Taylor
approximations around a steady state.
3. Note that when these models are linearized to the first order, agents
behave as if future shocks where equal to zero (since their expectation is
null), which is the certainty equivalence property. This is an often
overlooked point in the literature which misleads readers in supposing
their models may be deterministic.
3.2 Introducing an example
The goal of this first section is to introduce a simple example. Future sections
will aim to code this example into Dynare and analyze its salient features
under the influence of shocks - both in a stochastic and a deterministic envi-
ronment. Note that as a general rule, the examples in the basic chapters, 3
and 5, are kept as bare as possible, with just enough features to help illustrate
Dynare commands and functionalities. More complex examples are instead
presented in the advanced chapters.
The model introduced here is a basic RBC model with monopolistic com-
petition, used widely in the literature. Its particular notation adopted below
is drawn mostly from notes available on Jesus Fernandez-Villaverde’s very
instructive website; this is a good place to look for additional information
on any of the following model set-up and discussion. Note throughout this
model description that the use of expectation signs is really only relevant
in a stochastic setting, as per the earlier discussion. We will none-the-less
illustrate both the stochastic and the deterministic settings on the basis of
this example. Thus, when thinking of the latter, you’ll have to use a bit of
imagination (on top of that needed to think you have perfect foresight!) to
12 CHAPTER 3. SOLVING DSGE MODELS - BASICS
ignore the expectation signs.
Households maximize utility over consumption, c
t
and leisure, 1−l
t
, where
l
t
is labor input, according to the following utility function
E
t

t=0
β [log c
t
+ψ log(1 −l
t
)]
and subject to the following budget constraint
c
t
+k
t+1
= w
t
l
t
+r
t
k
t
+ (1 −δ)k
t
, ∀t > 0
where k
t
is capital stock, w
t
real wages, r
t
real interest rates or cost of capital
and δ the depreciation rate.
The above equation can be seen as an accounting identity, with total ex-
penditures on the left hand side and revenues - including the liquidation value
of the capital stock - on the right hand side. Alternatively, with a little more
imagination, the equation can also be interpreted as a capital accumulation
equation after bringing c
t
to the right hand side and noticing that w
t
l
t
+r
t
k
t
,
total payments to factors, equals y
t
, or aggregate output, by the zero profit
condition. As a consequence, if we define investment as i
t
= y
t
−c
t
, we obtain
the intuitive result that i
t
= k
t+1
− (1 − δ)k
t
, or that investment replenishes
the capital stock thereby countering the effects of depreciation. In any given
period, the consumer therefore faces a tradeoff between consuming and in-
vesting in order to increase the capital stock and consuming more in following
periods (as we will see later, production depends on capital).
Maximization of the household problem with respect to consumption,
leisure and capital stock, yields the Euler equation in consumption, capturing
the intertemporal tradeoff mentioned above, and the labor supply equation
linking labor positively to wages and negatively to consumption (the wealth-
ier, the more leisure due to the decreasing marginal utility of consumption).
These equation are
1
c
t
= βE
t
_
1
c
t+1
(1 +r
t+1
−δ)
_
and
ψ
c
t
1 −l
t
= w
t
The firm side of the problem is slightly more involved, due to monopolistic
competition, but is presented below in the simplest possible terms, with a
little hand-waiving involved, as the derivations are relatively standard.
3.2. INTRODUCING AN EXAMPLE 13
There are two ways to introduce monopolistic competition. We can ei-
ther assume that firms sell differentiated varieties of a good to consumers who
aggregate these according to a CES index. Or we can postulate that there
is a continuum of intermediate producers with market power who each sell
a different variety to a competitive final goods producer whose production
function is a CES aggregate of intermediate varieties.
If we follow the second route, the final goods producer chooses his or her
optimal demand for each variety, yielding the Dixit-Stiglitz downward sloping
demand curve. Intermediate producers, instead, face a two pronged decision:
how much labor and capital to employ given these factors’ perfectly competi-
tive prices and how to price the variety they produce.
Production of intermediate goods follows a CRS production function de-
fined as
y
it
= k
α
it
(e
z
t
l
it
)
1−α
where the i subscript stands for firm i of a continuum of firms between zero
and one and where α is the capital elasticity in the production function, with
0 < α < 1. Also, z
t
captures technology which evolves according to
z
t
= ρz
t−1
+e
t
where ρ is a parameter capturing the persistence of technological progress and
e
t
∼ N(0, σ).
The solution to the sourcing problem yields an optimal capital to labor
ratio, or relationship between payments to factors:
k
it
r
t
=
α
1 −α
w
t
l
it
The solution to the pricing problem, instead, yields the well-known con-
stant markup pricing condition of monopolistic competition:
p
it
=

−1
mc
t
p
t
where p
it
is firm i’s specific price, mc
t
is real marginal cost and p
t
is the aggre-
gate CES price or average price. An additional step simplifies this expression:
symmetric firms implies that all firms charge the same price and thus p
it
= p
t
;
we therefore have: mc
t
= ( −1)/
But what are marginal costs equal to? To find the answer, we combine the
optimal capital to labor ratio into the production function and take advantage
of its CRS property to solve for the amount of labor or capital required to
produce one unit of output. The real cost of using this amount of any one
14 CHAPTER 3. SOLVING DSGE MODELS - BASICS
factor is given by w
t
l
it
+ r
t
k
it
where we substitute out the payments to the
other factor using again the optimal capital to labor ratio. When solving for
labor, for instance, we obtain
mc
t
=
_
1
1 −α
_
1−α
_
1
α
_
α
1
A
t
w
1−α
t
r
α
t
which does not depend on i; it is thus the same for all firms.
Interestingly, the above can be worked out, by using the optimal capital
to labor ratio, to yield w
t
[(1 − α)y
it
/l
it
]
−1
, or w
t
∂l
it
∂y
it
, which is the definition
of marginal cost: the cost in terms of labor input of producing an additional
unit of output. This should not be a surprise since the optimal capital to
labor ratio follows from the maximization of the production function (minus
real costs) with respect to its factors.
Combining this result for marginal cost, as well as its counterpart in terms
of capital, with the optimal pricing condition yields the final two important
equations of our model
w
t
= (1 −α)
y
it
l
it
( −1)

and
r
t
= α
y
it
k
it
( −1)

To end, we aggregate the production of each individual firm to find an
aggregate production function. On the supply side, we factor out the capital
to labor ratio, k
t
/l
t
, which is the same for all firms and thus does not depend
on i. On the other side, we have the Dixit-Stiglitz demand for each variety. By
equating the two and integrating both side, and noting that price dispersion
is null - or that, as hinted earlier, p
it
= p
t
- we obtain aggregate production
y
t
= A
t
k
α
t
l
1−α
t
which can be shown is equal to the aggregate amount of varieties bought by
the final good producer (according to a CES aggregation index) and, in turn,
equal to the aggregate output of final good, itself equal to household con-
sumption. Note, to close, that because the ratio of output to each factor is
the same for each intermediate firm and that firm specific as well as aggre-
gate production is CRS, we can rewrite the above two equations for w
t
and r
t
without the i subscripts on the right hand side.
This ends the exposition of the example. Now, let’s roll up our sleeves and
see how we can input the model into Dynare and actually test how the model
will respond to shocks.
3.3. DYNARE .MOD FILE STRUCTURE 15
3.3 Dynare .mod file structure
Input into Dynare involves the .mod file, as mentioned loosely in the intro-
duction of this Guide. The .mod file can be written in any editor, external or
internal to Matlab. It will then be read by Matlab by first navigating within
Matlab to the directory where the .mod file is stored and then by typing in
the Matlab command line Dynare filename.mod; (although actually typing
the extension .mod is not necessary). But before we get into executing a .mod
file, let’s start by writing one!
It is convenient to think of the .mod file as containing four distinct blocks,
illustrated in figure 3.3:
• preamble: lists variables and parameters
• model: spells out the model
• steady state or initial value: gives indications to find the steady state
of a model, or the starting point for simulations or impulse response
functions based on the model’s solution.
• shocks: defines the shocks to the system
• computation: instructs Dynare to undertake specific operations (e.g.
forecasting, estimating impulse response functions)
Our exposition below will structured according to each of these blocks.
3.4 Filling out the preamble
The preamble generally involves three commands that tell Dynare what are
the model’s variables, which are endogenous and what are the parameters.
The commands are:
• var starts the list of endogenous variables, to be separated by commas.
• varexo starts the list of exogenous variables that will be shocked.
• parameters starts the list of parameters and assigns values to each.
In the case of our example, let’s differentiate between the stochastic and de-
terministic cases. First, we lay these out, then we discuss them.
16 CHAPTER 3. SOLVING DSGE MODELS - BASICS
Figure 3.1: The .mod file contains five logically distinct parts.
3.4.1 The deterministic case
The model is inherited exactly as specified in the earlier description, except
that we no longer need the e
t
variable, as we can make z
t
directly exogenous.
Thus, the preamble would look like:
var y c k i l y l w r;
varexo z;
parameters beta psi delta alpha sigma epsilon;
alpha = 0.33;
beta = 0.99;
delta = 0.023;
psi = 1.75;
sigma = (0.007/(1-alpha));
epsilon = 10;
3.4.2 The stochastic case
In this case, we go back to considering the law of motion for technology, con-
sisting of an exogenous shock, e
t
. With respect to the above, we therefore
3.4. FILLING OUT THE PREAMBLE 17
adjust the list of endogenous and exogenous variables, and add the parameter
ρ. Here’s what the preamble would look like:
var y c k i l y l w r z;
varexo e;
parameters beta psi delta alpha rho sigma epsilon;
alpha = 0.33;
beta = 0.99;
delta = 0.023;
psi = 1.75;
rho = 0.95;
sigma = (0.007/(1-alpha));
epsilon = 10;
3.4.3 Comments on your first lines of Dynare code
As you can tell, writing a .mod file is really quite straightforward. Two quick
comments:
NOTE! Remember that each instruction of the .mod file must be termi-
nated by a semicolon (;), although a single instruction can span two lines if
you need extra space (just don’t put a semicolon at the end of the first line).
TIP! You can also comment out any line by starting the line with two
forward slashes (//), or comment out an entire section by starting the section
with /* and ending with */. For example:
var y c k i l y l w r z;
varexo e;
parameters beta psi delta
alpha rho sigma epsilon;
// the above instruction reads over two lines
/*
the following section lists
several parameters which were
calibrated by my co-author. Ask
her all the difficult questions!
*/
alpha = 0.33;
beta = 0.99;
delta = 0.023;
psi = 1.75;
rho = 0.95;
18 CHAPTER 3. SOLVING DSGE MODELS - BASICS
sigma = (0.007/(1-alpha));
epsilon = 10;
3.5 Specifying the model
3.5.1 Model in Dynare notation
One of the beauties of Dynare is that you can input your model’s equa-
tions naturally, almost as if you were writing them in an academic paper.
This greatly facilitates the sharing of your Dynare files, as your colleagues will
be able to understand your code in no-time. There are just a few conventions
to follow. Let’s first have a look at our model in Dynare notation, and
then go through the various Dynare input conventions. What you can already
try to do is glance at the model block below and see if you can recognize the
equations from the earlier example. See how easy it is to read Dynare code?
model;
(1/c) = beta*(1/c(+1))*(1+r(+1)-delta);
psi*c/(1-l) = w;
c+i = y;
y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha);
w = y*((epsilon-1)/epsilon)*(1-alpha)/l;
r = y*((epsilon-1)/epsilon)*alpha/k(-1);
i = k-(1-delta)*k(-1);
y l = y/l;
z = rho*z(-1)+e;
end;
Just in case you need a hint or two to recognize these equations, here’s
a brief description: the first equation is the Euler equation in consumption.
The second the labor supply function. The third the accounting identity. The
fourth is the production function. The fifth and sixth are the marginal cost
equal to markup equations. The seventh is the investment equality. The
eighth an identity that may be useful and the last the equation of motion of
technology.
NOTE! that the above model specification corresponds to the stochastic
case; indeed, notice that the law of motion for technology is included, as per
our discussion of the preamble. The corresponding model for the determin-
istic casce would simply loose the last equation.
3.5. SPECIFYING THE MODEL 19
3.5.2 General conventions
The above example illustrates the use of a few important commands and
conventions to translate a model into a Dynare-readable .mod file.
• The first thing to notice, is that the model block of the .mod file begins
with the command model and ends with the command end.
• Second, in between, there need to be as many equations as you declared
endogenous variables (this is actually one of the first things that Dynare
checks; it will immediately let you know if there are any problems).
• Third, as in the preamble and everywhere along the .mod file, each line
of instruction ends with a semicolon (except when a line is too long and
you want to break it across two lines. This is unlike Matlab where if you
break a line you need to add . . . ).
• Fourth, equations are entered one after the other; no matrix representa-
tion is necessary. Note that variable and parameter names used in the
model block must be the same as those declared in the preamble; TIP!
remember that variable and parameter names are case sensitive.
3.5.3 Notational conventions
• Variables entering the system with a time t subscript are written plainly.
For example, x
t
would be written x.
• Variables entering the system with a time t − n subscript are written
with (−n) following them. For example, x
t−2
would be written x(−2)
(incidentally, this would count as two backward looking variables).
• In the same way, variables entering the system with a time t+n subscript
are written with (+n) following them. For example, x
t+2
would be
written x(+2). Writing x(2) is also allowed, but this notation makes it
slightly harder to count by hand the number of forward looking variables
(a useful measure to check); more on this below . . .
3.5.4 Timing conventions
• In Dynare, the timing of each variable reflects when that variable is de-
cided. For instance, our capital stock is not decided today, but yesterday
(recall that it is a function of yesterday’s investment and capital stock);
it is what we call in the jargon a predetermined variable. Thus, even-
though in the example presented above we wrote k
t+1
= i
t
+ (1 − δ)k
t
,
as in many papers, we would translate this equation into Dynare as
k=i+(1-delta)*k(-1).
20 CHAPTER 3. SOLVING DSGE MODELS - BASICS
• As another example, consider that in some wage negociation models,
wages used during a period are set the period before. Thus, in the
equation for wages, you can write wage in period t (when they are set),
but in the labor demand equation, wages should appear with a one
period lag.
• A slightly more roundabout way to explain the same thing is that for
stock variables, you must use a “stock at the end of the period” concept.
It is investment during period t that sets stock at the end of period t.
Be careful, a lot of papers use the “stock at the beginning of the period”
convention, as we did (on purpose to highlight this distinction!) in the
setup of the example model above.
3.5.5 Conventions specifying non-predetermined variables
• A (+1) next to a variable tells Dynare to count the occurrence of that
variable as a jumper or forward-looking or non-predetermined variable.
• Blanchard-Kahn conditions are met only if the number of non-predetermined
variables equals the number of eigenvalues greater than one. If this con-
dition is not met, Dynare will put up a warning.
• Note that a variable may occur both as predetermined and non-predetermined.
For instance, consumption could appear with a lead in the Euler equa-
tion, but also with a lag in a habit formation equation, if you had one.
In this case, the second order difference equation would have two eigen-
values, one needing to be greater and the other smaller than one for
stability.
3.5.6 Linear and log-linearized models
There are two other variants of the system’s equations which Dynare accom-
modates. First, the linear model and second, the model in exp-logs. In
the first case, all that is necessary is to write the term (linear) next to the
command model. Our example, with just the equation for y
l
for illustration,
would look like:
model (linear);
yy l=yy - ll;
end;
where repeating a letter for a variable means difference from steady state.
Otherwise, you may be interested to have Dynare take Taylor series ex-
pansions in logs rather than in levels; this turns out to be a very useful option
3.6. SPECIFYING STEADY STATES AND/OR INITIAL VALUES 21
when estimating models with unit roots, as we will see in chapter 5. If so,
simply rewrite your equations by taking the exponential and logarithm of each
variable. The Dynare input convention makes this very easy to do. Our ex-
ample would need to be re-written as follows (just shown for the first two
equations)
model;
(1/exp(cc)) = beta*(1/exp(cc(+1)))*(1+exp(rr(+1))-delta);
psi*exp(cc)/(1-exp(ll)) = exp(ww);
end;
where, this time, repeating a letter for a variable means log of that variable,
so that the level of a variable is given by exp(repeatedvariable).
3.6 Specifying steady states and/or initial values
Material in this section has created much confusion in the past. But with
some attention to the explanations below, you should get through unscathed.
Let’s start by emphasizing the uses of this section of the .mod file. First, recall
that stochastic models need to be linearized. Thus, they need to have a steady
state. One of the functions of this section is indeed to provide these steady
state values, or approximations of values. Second, irrespective of whether
you’re working with a stochastic or deterministic model, you may be inter-
ested to start your simulations or impulse response functions from either a
steady state, or another given point. This section is also useful to specify this
starting value. Let’s see in more details how all this works.
In passing, though, note that the relevant commands in this section are
initval, endval or, more rarely, histval which is covered only in the Ref-
erence Manual. The first two are instead covered in what follows.
3.6.1 Stochastic models and steady states
In a stochastic setting, your model will need to be linearized before it is solved.
To do so, Dynare needs to know your model’s steady state (more details on
finding a steady state, as well as tips to do so more efficiently, are provided in
section 3.6.3 below). You can either enter exact steady state values into your
.mod file, or just approximations and let Dynare find the exact steady state
(which it will do using numerical methods based on your approximations). In
either case, these values are entered in the initval block, as in the following
fashion:
initval;
22 CHAPTER 3. SOLVING DSGE MODELS - BASICS
k = 9;
c = 0.7;
l = 0.3;
w = 2.0;
r = 0;
z = 0;
e = 0;
end;
Then, by using the command steady, you can control whether you want to
start your simulations or impulse response functions from the steady state, or
from the exact values you specified in the initval block. Adding steady just
after your initval block will instruct Dynare to consider your initial values
as mere approximations and start simulations or impulse response functions
from the exact steady state. On the contrary, if you don’t add the command
steady, your simulations or impulse response functions will start from your
initial values, even if Dynare will have calculated your model’s exact steady
state for the purpose of linearization.
For the case in which you would like simulations and impulse response
functions to begin at the steady state, the above block would be expanded to
yield:
initval;
k = 9;
c = 0.7;
l = 0.3;
w = 2.0;
r = 0;
z = 0;
e = 0;
end;
steady;
TIP! If you’re dealing with a stochastic model, remember that its lin-
ear approximation is good only in the vicinity of the steady state, thus it is
strongly recommended that you start your simulations from a steady state;
this means either using the command steady or entering exact steady state
values.
3.6. SPECIFYING STEADY STATES AND/OR INITIAL VALUES 23
3.6.2 Deterministic models and initial values
Deterministic models do not need to be linearized in order to be solved. Thus,
technically, you do not need to provide a steady state for these model. But
practically, most researchers are still interested to see how a model reacts to
shocks when originally in steady state. In the deterministic case, the initval
block serves very similar functions as described above. If you wanted to shock
your model starting from a steady state value, you would enter approximate
(or exact) steady state values in the initval block, followed by the command
steady. Otherwise, if you wanted to begin your solution path from an arbi-
trary point, you would enter those values in your initval block and not use
the steady command. An illustration of the initval block in the determin-
istic case appears further below.
3.6.3 Finding a steady state
The difficulty in the above, of course, is calculating actual steady state val-
ues. Doing so borders on a form of art, and luck is unfortunately part of the
equation. Yet, the following TIPS! may help.
As mentioned above, Dynare can help in finding your model’s steady state
by calling the appropriate Matlab functions. But it is usually only successful
if the initial values you entered are close to the true steady state. If you have
trouble finding the steady state of your model, you can begin by playing with
the options following the steady command. These are:
• solve algo = 0: uses Matlab Optimization Toolbox FSOLVE
• solve algo = 1: uses Dynare’s own nonlinear equation solver
• solve algo = 2: splits the model into recursive blocks and solves each
block in turn.
• solve algo = 3: uses the Sims solver. This is the default option if none
are specified.
For complicated models, finding suitable initial values for the endogenous
variables is the trickiest part of finding the equilibrium of that model. Often,
it is better to start with a smaller model and add new variables one by one.
But even for simpler models, you may still run into difficulties in finding
your steady state. If so, another option is to enter your model in linear
terms. In this case, variables would be expressed in percent deviations from
steady state. Thus, their initial values would all be zero. Unfortunately, if
any of your original (non-linear) equations involve sums (a likely fact), your
24 CHAPTER 3. SOLVING DSGE MODELS - BASICS
linearized equations will include ratios of steady state values, which you would
still need to calculate. Yet, you may be left needing to calculate fewer steady
state values than in the original, non-linear, model.
Alternatively, you could also use an external program to calculate ex-
act steady state values. For instance, you could write an external Maple
file and then enter the steady state solution by hand in Dynare. But of
course, this procedure could be time consuming and bothersome, especially
if you want to alter parameter values (and thus steady states) to undertake
robustness checks.
The alternative is to write a Matlab program to find your model’s steady
state. Doing so has the clear advantages of being able to incorporate your
Matlab program directly into your .mod file so that running loops with differ-
ent parameter values, for instance, becomes seamless. NOTE! When doing so,
your matlab (.m) file should have the same name as your .mod file, followed
by steadystate For instance, if your .mod file is called example.mod, your
Matlab file should be called example steadystate.m and should be saved in
the same directory as your .mod file. Dynare will automatically check the di-
rectory where you’ve saved your .mod file to see if such a Matlab file exists. If
so, it will use that file to find steady state values regardless of whether you’ve
provided initial values in your .mod file.
Because Matlab does not work with analytical expressions, though (unless
you’re working with a particular toolbox), you need to do a little work to write
your steady state program. It is not enough to simply input the equations
as you’ve written them in your .mod file and ask Matlab to solve the system.
You will instead need to write your steady state program as if you were solv-
ing for the steady state by hand. That is, you need to input your expressions
sequentially, whereby each left-hand side variable is written in terms of known
parameters or variables already solved in the lines above. For example, the
steady state file corresponding to the above example, in the stochastic case,
would be: (** example file to be added shortly)
3.6.4 Checking system stability
TIP! A handy command that you can add after the initval or endval block
(following the steady command if you decide to add one) is the check com-
mand. This computes and displays the eigenvalues of your system
which are used in the solution method. As mentioned earlier, a necessary con-
dition for the uniqueness of a stable equilibrium in the neighborhood of the
steady state is that there are as many eigenvalues larger than one in modulus
as there are forward looking variables in the system. If this condition is not
3.7. ADDING SHOCKS 25
met, Dynare will tell you that the Blanchard-Kahn conditions are not satisfied
(whether or not you insert the check command).
3.7 Adding shocks
3.7.1 Deterministic models - temporary shocks
When working with a deterministic model, you have the choice of introducing
both temporary and permanent shocks. The distinction is that under a tem-
porary shock, the model eventually comes back to steady state, while under
a permanent shock, the model reaches a new steady state. In both cases,
though, the shocks are entirely expected, as explained in our original discus-
sion on stochastic and deterministic models.
To work with a temporary shock, you are free to set the duration and
level of the shock. To specify a shock that lasts 9 periods on z
t
, for instance,
you would write:
shocks;
var z;
periods 1:9;
values 0.1;
end;
Given the above instructions, Dynare would replace the value of z
t
spec-
ified in the initval block with the value of 0.1 entered above. If variables
were in logs, this would have corresponded to a 10% shock. Note that you
can also use the mshocks command which multiplies the initial value of an
exogenous variable by the mshocks value. Finally, note that we could have
entered future periods in the shocks block, such as periods 5:10, in order to
study the anticipatory behavior of agents in response to future shocks.
3.7.2 Deterministic models - permanent shocks
To study the effects of a permanent shock hitting the economy today, such
as a structural change in your model, you would not specify actual “shocks”,
but would simply tell the system to which (steady state) values you would like
it to move and let Dynare calculate the transition path. To do so, you would
use the endval block following the usual initval block. For instance, you
may specify all values to remain common between the two blocks, except for
the value of technology which you may presume changes permanently. The
corresponding instructions would be:
26 CHAPTER 3. SOLVING DSGE MODELS - BASICS
initval;
k = 9;
c = 0.7;
l = 0.3;
w = 2.0;
r = 0;
z = 0;
end;
steady;
endval;
k = 9;
c = 0.7;
l = 0.3;
w = 2.0;
r = 0;
z = 0.1;
end;
steady;
where steady can also be added to the endval block, and serves the same
functionality as described earlier (namely, of telling Dynare to start and/ or
end at a steady state close to the values you entered. If you do not use steady
after endval, and the latter does not list exact steady state values, you may
impose on your system that it does not return to steady state. This is unusual.
In this case, your problem would become a so-called two boundary problem,
which, when solved, requires that the path of your endogenous variables pass
through the steady state closest to your endval values). In our example, we
make use of the second steady since the actual terminal steady state values
are bound to be somewhat different from those entered above, which are noth-
ing but the initial values for all variables except for technology.
In the above example, the value of technology would move to 0.1 in pe-
riod 1 (tomorrow) and thereafter. But of course, the other variables - the
endogenous variables - will take longer to reach their new steady state values.
TIP! If you instead wanted to study the effects of a permanent but future
shock (anticipated as usual), you would have to add a shocks block after the
endval block to “undo” the first several periods of the permanent shock. For
instance, suppose you wanted the value of technology to move to 0.1, but only
in period 10. Then you would follow the above endval block with:
shocks;
var z;
3.8. SELECTING A COMPUTATION 27
periods 1:9;
values 0;
end;
3.7.3 Stochastic models
Recall from our earlier description of stochastic models that shocks are only
allowed to be temporary. A permanent shock cannot be accommodated due to
the need to stationarize the model around a steady state. Furthermore, shocks
can only hit the system today, as the expectation of future shocks must be
zero. With that in mind, we can however make the effect of the shock propa-
gate slowly throughout the economy by introducing a “latent shock variable”
such as e
t
in our example, that affects the model’s true exogenous variable, z
t
in our example, which is itself an AR(1), exactly as in the model we introduced
from the outset. In that case, though, we would declare z
t
as an endogenous
variable and e
t
as an exogenous variable, as we did in the preamble of the
.mod file in section 3.4. Supposing we wanted to add a shock with variance
σ
2
, where σ is determined in the preamble block, we would write:
shocks;
var e = sigma^2;
end;
TIP! You can actually mix in deterministic shocks in stochastic models
by using the commands varexo det and listing some shocks as lasting more
than one period in the shocks block. For information on how to do so, please
see the Reference Manual. This can be particularly useful if you’re studying
the effects of anticipated shocks in a stochastic model. For instance, you may
be interested in what happens to your monetary model if agents began ex-
pecting higher inflation, or a depreciation of your currency.
3.8 Selecting a computation
So far, we have written an instructive .mod file, but what should Dynare do
with it? What are we interested in? In most cases, it will be impulse re-
sponse functions (IRFs) due to the external shocks. Let’s see which are the
appropriate commands to give to Dynare. Again, we will distinguish between
deterministic and stochastic models.
28 CHAPTER 3. SOLVING DSGE MODELS - BASICS
3.8.1 For deterministic models
In the deterministic case, all you need to do is add the command simul
at the bottom of your .mod file. Note that the command takes the option
[ (periods=INTEGER) ] The command simul triggers the computation a
numerical simulation of the trajectory of the model’s solution for the number
of periods set in the option. To do so, it uses a Newton method to solve
simultaneously all the equations for every period (see Juillard (1996) for de-
tails). Note that unless you use the endval command, the algorithm makes
the simplifying assumption that the system is back to equilibrium after the
specified number of periods. Thus, you must specify a large enough number
of periods such that increasing it further doesn’t change the simulation for
all practical purpose. In the case of a temporary shock, for instance, the tra-
jectory will basicaly describe how the system gets back to equilibrium after
being perturbed from the shocks you entered.
3.8.2 For stochastic models
In the more common case of stochastic models, the command stoch simul is
appropriate. This command instructs Dynare to compute a Taylor approxi-
mation of the decision and transition functions for the model (the equations
listing current values of the endogenous variables of the model as a func-
tion of the previous state of the model and current shocks), impulse response
functions and various descriptive statistics (moments, variance decomposition,
correlation and autocorrelation coefficients).
1
Impulse response functions are the expected future path of the endogenous
variables conditional on a shock in period 1 of one standard deviation.TIP!
If you linearize your model up to a first order, impulse response functions
are simply the algebraic forward iteration of your model’s policy or decision
rule. If you instead linearize to a second order, impulse response functions
will be the result of actual Monte Carlo simulations of future shocks. This is
because in second order linear equations, you will have cross terms involving
the shocks, so that the effects of the shocks depend on the state of the system
when the shocks hit. Thus, it is impossible to get algebraic average values
of all future shocks and their impact. The technique is instead to pull fu-
ture shocks from their distribution and see how they impact your system, and
repeat this procedure a multitude of times in order to draw out an average
response. That said, note that future shocks will not have a significant impact
1
For correlated shocks, the variance decomposition is computed as in the VAR literature
through a Cholesky decomposition of the covariance matrix of the exogenous variables.
When the shocks are correlated, the variance decomposition depends upon the order of the
variables in the varexo command.
3.8. SELECTING A COMPUTATION 29
on your results, since they get averaged between each Monte Carlo trial and
in the limit should sum to zero, given their mean of zero. Note that in the
case of a second order approximation, Dynare will return the actual sample
moments from the simulations. For first order linearizations, Dynare will in-
stead report theoretical moments. In both cases, the return to steady state
is asymptotic, TIP! thus you should make sure to specify sufficient periods
in your IRFs such that you actually see your graphs return to steady state.
Details on implementing this appear below.
If you’re interested to peer a little further into what exactly is going on
behind the scenes of Dynare’s computations, have a look at Chapter 7. Here
instead, we focus on the application of the command and reproduce below the
most common options that can be added to stoch simul. For a complete list
of options, please see the Reference Manual.
Options following the stoch simul command:
• ar = INTEGER: Order of autocorrelation coefficients to compute and
to print (default = 5).
• dr algo = 0 or 1: specifies the algorithm used for computing the quadratic
approximation of the decision rules: 0 uses a pure perturbation approach
as in Schmitt-Grohe and Uribe (2004) (default) and 1 moves the point
around which the Taylor expansion is computed toward the means of
the distribution as in Collard and Juillard (2001b).
• drop = INTEGER: number of points dropped at the beginning of sim-
ulation before computing the summary statistics (default = 100).
• hp filter = INTEGER: uses HP filter with lambda = INTEGER before
computing moments (default: no filter).
• hp ngrid = INTEGER: number of points in the grid for the discreet In-
verse Fast Fourier Transform used in the HP filter computation. It may
be necessary to increase it for highly autocorrelated processes (default
= 512).
• irf = INTEGER: number of periods on which to compute the IRFs
(default = 40). Setting IRF=0, suppresses the plotting of IRF’s.
• relative irf requests the computation of normalized IRFs in percentage
of the standard error of each shock.
• nocorr: doesn’t print the correlation matrix (printing is the default).
• nofunctions: doesn’t print the coefficients of the approximated solution
(printing is the default).
30 CHAPTER 3. SOLVING DSGE MODELS - BASICS
• nomoments: doesn’t print moments of the endogenous variables (print-
ing them is the default).
• noprint: cancel any printing; usefull for loops.
• order = 1 or 2 : order of Taylor approximation (default = 2), unless
you’re working with a linear model in which case the order is automati-
cally set to 1.
• periods = INTEGER: specifies the number of periods to use in simu-
lations (default = 0). TIP! A simulation is similar to running impulse
response functions with a model linearized to the second order, in the
way that both sample shocks from their distribution to see how the
system reacts, but a simulation only repeats the process once, whereas
impulse response functions run a multitude of Monte Carlo trials in order
to get an average response of your system.
• qz criterium = INTEGER or DOUBLE: value used to split stable from
unstable eigenvalues in reordering the Generalized Schur decomposition
used for solving 1st order problems (default = 1.000001).
• replic = INTEGER: number of simulated series used to compute the
IRFs (default = 1 if order = 1, and 50 otherwise).
• simul seed = INTEGER or DOUBLE or (EXPRESSION): specifies a
seed for the random number generator so as to obtain the same random
sample at each run of the program. Otherwise a different sample is
used for each run (default: seed not specified). If you linearized to a
second order, Dynare will actually undertake Monte Carlo simulations
to generate moments of your variables. Because of the simulation, results
are bound to be slightly different each time you run your program, except
if you fix the seed for the random number generator. TIP! If you do
decide to fix the seed, you should at least try to run your program
without using simul seed, just to check the robustness of your results.
Going back to our good old example, suppose we were interested in print-
ing all the various measures of moments of our variables, want to see impulse
response functions for all variables, are basically happy with all default op-
tions and want to carry out simulations over a good number of periods. We
would then end our .mod file with the following command:
stoch simul(periods=2100);
3.9. THE COMPLETE .MOD FILE 31
3.9 The complete .mod file
For completion’s sake, and for the pleasure of seeing our work bear its fruits,
here are the complete .mod files corresponding to our example for the de-
terministic and stochastic case. You can find the corresponding files in the
models folder under UserGuide in your installation of Dynare. The files are
called RBC Monop JFV.mod for stochastic models and RBC Monop Det.mod for
deterministic models.
3.9.1 The stochastic model
var y c k i l y l w r z;
varexo e;
parameters beta psi delta alpha rho gamma sigma epsilon;
alpha = 0.33;
beta = 0.99;
delta = 0.023;
psi = 1.75;
rho = 0.95;
sigma = (0.007/(1-alpha));
epsilon = 10;
model;
(1/c) = beta*(1/c(+1))*(1+r(+1)-delta);
psi*c/(1-l) = w;
c+i = y;
y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha);
w = y*((epsilon-1)/epsilon)*(1-alpha)/l;
r = y*((epsilon-1)/epsilon)*alpha/k(-1);
i = k-(1-delta)*k(-1);
y l = y/l;
z = rho*z(-1)+e;
end;
initval;
k = 9;
c = 0.76;
l = 0.3;
w = 2.07;
r = 0.03;
z = 0;
e = 0;
end;
32 CHAPTER 3. SOLVING DSGE MODELS - BASICS
steady;
check;
shocks;
var e = sigma^2;
end;
stoch simul(periods=2100);
3.9.2 The deterministic model (case of temporary shock)
var y c k i l y l w r ;
varexo z;
parameters beta psi delta alpha sigma epsilon;
alpha = 0.33;
beta = 0.99;
delta = 0.023;
psi = 1.75;
sigma = (0.007/(1-alpha));
epsilon = 10;
model;
(1/c) = beta*(1/c(+1))*(1+r(+1)-delta);
psi*c/(1-l) = w;
c+i = y;
y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha);
w = y*((epsilon-1)/epsilon)*(1-alpha)/l;
r = y*((epsilon-1)/epsilon)*alpha/k(-1);
i = k-(1-delta)*k(-1);
y l = y/l;
end;
initval;
k = 9;
c = 0.7;
l = 0.3;
w = 2.0;
r = 0;
z = 0;
end;
steady;
3.10. FILE EXECUTION AND RESULTS 33
check;
shocks;
var z;
periods 1:9;
values 0.1;
end;
simul(periods=2100);
3.10 File execution and results
To see this all come to life, let’s run our .mod file, which is conveniently
installed by default in the Dynare “examples” directory (the .mod file cor-
responding to the stochastic model is called RBC Monop JFV.mod and that
corresponding to the deterministic model is called RBC Monop Det.mod). (**
note, this may not be the case when testing the beta version of Matlab version
4)
To run a .mod file, navigate within Matlab to the directory where the
example .mod files are stored. You can do this by clicking in the “current di-
rectory” window of Matlab, or typing the path directly in the top white field
of Matlab. Once there, all you need to do is place your cursor in the Matlab
command window and type, for instance, dynare ExSolStoch; to execute
your .mod file.
Running these .mod files should take at most 30 seconds. As a result, you
should get two forms of output - tabular in the Matlab command window and
graphical in one or more pop-up windows. Let’s review these results.
3.10.1 Results - stochastic models
The tabular results can be summarized as follows:
1. Model summary: a count of the various variable types in your model
(endogenous, jumpers, etc...).
2. Eigenvalues should be displayed, and you should see a confirmation of
the Blanchard-Kahn conditions if you used the command check in your
.mod file.
34 CHAPTER 3. SOLVING DSGE MODELS - BASICS
3. Matrix of covariance of exogenous shocks: this should square with
the values of the shock variances and co-variances you provided in the
.mod file.
4. Policy and transition functions: Solving the rational exectation
model, E
t
[f(y
t+1
, y
t
, y
t−1
, u
t
)] = 0 , means finding an unkown function,
y
t
= g(y
t−1
, u
t
) that could be plugged into the original model and satisfy
the implied restrictions (the first order conditions). A first order approx-
imation of this function can be written as y
t
= ¯ y + g
y
ˆ y
t−1
+ g
u
u
t
, with
ˆ y
t
= y
t
− ¯ y and ¯ y being the steadystate value of y, and where g
x
is the
partial derivative of the g function with respect to variable x. In other
words, the function g is a time recursive (approximated) representation
of the model that can generate timeseries that will approximatively sat-
isfy the rational expectation hypothesis contained in the original model.
In Dynare, the table “Policy and Transition function” contains the el-
ements of g
y
and g
u
. Details on the policy and transition function can
be found in Chapter 6.
5. Moments of simulated variables: up to the fourth moments.
6. Correlation of simulated variables: these are the contemporaneous
correlations, presented in a table.
7. Autocorrelation of simulated variables: up to the fifth lag, as spec-
ified in the options of stoch simul.
The graphical results, instead, show the actual impulse response func-
tions for each of the endogenous variables, given that they actually moved.
These can be especially useful in visualizing the shape of the transition func-
tions and the extent to which each variable is affected. TIP! If some variables
do not return to their steady state, either check that you have included enough
periods in your simulations, or make sure that your model is stationary, i.e.
that your steady state actually exists and is stable. If not, you should detrend
your variables and rewrite your model in terms of those variables.
3.10.2 Results - deterministic models
Automatically displayed results are much more scarce in the case of deter-
ministic models. If you entered steady, you will get a list of your steady
state results. If you entered check, eigenvalues will also be displayed and you
should receive a statement that the rank condition has been satisfied, if all
goes well! Finally, you will see some intermediate output: the errors at each
iteration of the Newton solver used to estimate the solution to your model.
TIP! You should see these errors decrease upon each iteration; if not, your
model will probably not converge. If so, you may want to try to increase the
periods for the transition to the new steady state (the number of simulations
3.10. FILE EXECUTION AND RESULTS 35
periods). But more often, it may be a good idea to revise your equations. Of
course, although Dynare does not display a rich set of statistics and graphs
corresponding to the simulated output, it does not mean that you cannot cre-
ate these by hand from Matlab. To do so, you should start by looking at
section 4.1.3 of chapter 4 on finding, saving and viewing your output.
Chapter 4
Solving DSGE models -
advanced topics
This chapter is a collection of topics - not all related to each other - that you
will probably find interesting or at least understandable, if you have read,
and/ or feel comfortable with, the earlier chapter 3 on the basics of solving
DSGE models. To provide at least some consistency, this chapter is divided
into three sections. The first section deals directly with features of Dynare,
such as dealing with correlated shocks, finding and saving your output, using
loops, referring to external files and dealing with infinite eigenvalues. The
second section overviews some of the inner workings of Dynare. The goal
is to provide a brief explanation of the files that are created by Dynare to
help you in troubleshooting or provide a starting point in case you actually
want to customize the way Dynare works. The third section of the chapter
focusses on modeling tips optimized for Dynare, but possibly also helpful for
other work.
4.1 Dynare features and functionality
4.1.1 Other examples
Other examples of .mod files used to generate impulse response functions are
available on the Dynare website. In particular, Jesus Fernandez-Villaverde
has provided a series of RBC model variants (from the most basic to some
including variable capacity utilization, indivisible labor and investment spe-
cific technological change). You can find these, along with helpful notes and
explanations, in the Official Examples section of the Dynare website.
Also, don’t forget to check occasionally the Dynare contributions and ex-
amples forum to see if any other user has posted an example that could help
37
38 CHAPTER 4. SOLVING DSGE MODELS - ADVANCED TOPICS
you in your work; or maybe you would like to post an example there yourself?
4.1.2 Alternative, complete example
The following example aims to give you an alternative example to the one in
chapter 3, to learn the workings of Dynare. It also aims to give you exposure
to dealing with several correlated shocks. Your model may have two or
more shocks, and these may be correlated to each other. The example below
illustrates how you would introduce this into Dynare. Actually, the example
provided is somewhat more complete than strictly necessary. This is to give
you an alternative, full-blown example to the one described in chapter 3.
The model
The model is a simplified standard RBC model taken from Collard and Juil-
lard (2003) which served as the original User Guide for Dynare.
The economy consists of an infinitely living representative agent who values
consumption c
t
and labor services h
t
according to the following utility function
E
t

τ=t
β
τ−t
_
log(c
t
) −θ
h
1+ψ
t
1 +ψ
_
where, as usual, the discount factor 0 < β < 1, the disutility of labor θ > 0
and the labor supply elasticity ψ ≥ 0.
A social planner maximizes this utility function subject to the resource
constraint
c
t
+i
t
= y
t
where i
t
is investment and y
t
output. Consumers are therefore also owners
of the firms. The economy is a real economy, where part of output can be
consumed and part invested to form physical capital. As is standard, the law
of motion of capital is given by
k
t+1
= exp(b
t
)i
t
+ (1 −δ)k
t
with 0 < δ < 1, where δ is physical depreciation and b
t
a shock affecting
incorporated technological progress.
We assume output is produced according to a standard constant returns
to scale technology of the form
y
t
= exp(a
t
)k
α
t
h
1−α
t
with α being the capital elasticity in the production function, with 0 < α < 1,
and where a
t
represents a stochastic technological shock (or Solow residual).
4.1. DYNARE FEATURES AND FUNCTIONALITY 39
Finally, we specify a shock structure that allows for shocks to display
persistence across time and correlation in the current period. That is
_
a
t
b
t
_
=
_
ρ τ
τ ρ
__
a
t−1
b
t−1
_
+
_

t
ν
t
_
where |ρ + τ| < 1 and |ρ −τ| < 1 to ensure stationarity (we call ρ the coeffi-
cient of persistence and τ that of cross-persistence). Furthermore, we assume
E
t
(
t
) = 0, E
t

t
) = 0 and that the contemporaneous variance-covariance
matrix of the innovations
t
and ν
t
is given by
_
σ
2

ψσ

σ
ν
ψσ

σ
ν
σ
2
ν
_
and where corr(
t
ν
s
) = 0, corr(
t

s
) = 0 and corr(ν
t
ν
s
) = 0 for all t = s.
This system - probably quite similar to standard RBC models you have run
into - yields the following first order conditions (which are straightforward to
reproduce in case you have doubts. . . ) and equilibrium conditions drawn from
the description above. Note that the first equation captures the labor supply
function and the second the intertemporal consumption Euler equation.
c
t
θh
1+ψ
t
= (1 −α)y
t
1 = βE
t
__
exp(b
t
)c
t
exp(b
t+1
)c
t+1
__
exp(b
t+1

y
t+1
k
t+1
+ 1 −δ
__
y
t
= exp(a
t
)k
α
t
h
1−α
t
k
t+1
= exp(b
t
)i
t
+ (1 −δ)k
t
a
t
= ρa
t−1
+τb
t−1
+
t
b
t
= τa
t−1
+ρb
t−1

t
The .mod file
To “translate” the model into a language understandable by Dynare, we would
follow the steps outlined in chapter 3. We will assume that you’re comfort-
able with these and simply present the final .mod file below. Fist, though,
note that to introduce shocks into Dynare, we have two options (this was not
discussed in the earlier chapter). Either write:
shocks;
var e; stderr 0.009;
var u; stderr 0.009;
var e, u = phi*0.009*0.009;
end;
40 CHAPTER 4. SOLVING DSGE MODELS - ADVANCED TOPICS
where the last line specifies the contemporaneous correlation between our two
exogenous variables.
Alternatively, you can also write:
shocks;
var e = 0.009^2;
var u = 0.009^2;
var e, u = phi*0.009*0.009;
end;
So that you can gain experience by manipulating the entire model, here is
the complete .mod file corresponding to the above example. You can find the
corresponding file in the models folder under UserGuide in your installation
of Dynare. The file is called Alt Ex1.mod.
var y, c, k, a, h, b;
varexo e, u;
parameters beta, rho, alpha, delta, theta, psi, tau;
alpha = 0.36;
rho = 0.95;
tau = 0.025;
beta = 0.99;
delta = 0.025;
psi = 0;
theta = 2.95;
phi = 0.1;
model;
c*theta*h^(1+psi)=(1-alpha)*y;
k = beta*(((exp(b)*c)/(exp(b(+1))*c(+1)))
*(exp(b(+1))*alpha*y(+1)+(1-delta)*k));
y = exp(a)*(k(-1)^alpha)*(h^(1-alpha));
k = exp(b)*(y-c)+(1-delta)*k(-1);
a = rho*a(-1)+tau*b(-1) + e;
b = tau*a(-1)+rho*b(-1) + u;
end;
initval;
y = 1.08068253095672;
c = 0.80359242014163;
4.1. DYNARE FEATURES AND FUNCTIONALITY 41
h = 0.29175631001732;
k = 5;
a = 0;
b = 0;
e = 0;
u = 0;
end;
shocks;
var e; stderr 0.009;
var u; stderr 0.009;
var e, u = phi*0.009*0.009;
end;
stoch simul(periods=2100);
4.1.3 Finding, saving and viewing your output
Where is output stored? Most of the moments of interest are stored in global
variable oo You can easily browse this global variable in Matlab by either
calling it in the command line, or using the workspace interface. In global
variable oo you will find the following (NOTE! variables will always appear
in the order in which you declared them in the preamble block of your .mod
file):
• steady state: the steady state of your variables
• mean: the mean of your variables
• var: the variance of your variables
• autocorr: the various autocorrelation matrices of your variables. Each
row of these matrices will correspond to a variables in time t, and
columns correspond to the variables lagged 1, for the first matrix, then
lagged 2 for the second matrix, and so on. Thus, the matrix of auto-
correlations that is automatically displayed in the results after running
stoch simul has, running down each column, the diagonal elements of
each of the various autocorrelation matrices described here.
• gamma y: the matrices of autocovariances. gamma y{1} represents vari-
ances, while gamma y{2} represents autocovariances where variables on
each column are lagged by one period and so on. By default, Dynare will
return autocovariances with a lag of 5. The last matrix (gamma y{7} in
42 CHAPTER 4. SOLVING DSGE MODELS - ADVANCED TOPICS
the default case) returns the variance decomposition, where each col-
umn captures the independent contribution of each shock to the variance
of each variable.
Furthermore, if you decide to run impulse response functions, you will
find a global variable oo .irfs comprising of vectors named endogenous
variable exogenous variable, like y e, reporting the values of the endoge-
nous variables corresponding to the impulse response functions, as a result of
the independent impulse of each exogenous shock.
To save your simulated variables, you can add the following command at
the end of your .mod file: dynasave (FILENAME) [variable names separated
by commas] If no variable names are specified in the optional field, Dynare will
save all endogenous variables. In Matlab, variables saved with the dynasave
command can be retrieved by using the Matlab command load -mat FILENAME.
4.1.4 Referring to external files
You may find it convenient to refer to an external file, either to compute the
steady state of your model, or when specifying shocks in an external file. The
former is described in section 3.6 of chapter 3 when discussing steady states.
The advantage of using Matlab, say, to find your model’s steady state was
clear with respect to Dynare version 3, as the latter resorted to numerical
approximations to find steady state values. But Dynare version 4 now uses
the same analytical methods available in Matlab. For most usage scenarios,
you should therefore do just as well to ask Dynare to compute your model’s
steady state (except, maybe, if you want to run loops, to vary your parameter
values, for instance, in which case writing a Matlab program may be more
handy).
But you may also be interested in the second possibility described above,
namely of specifying shocks in an external file, to simulate a model based on
shocks from a prior estimation, for instance. You could then retrieve the ex-
ogenous shocks from the oo file by saving them in a file called datafile.mat.
Finally, you could simulate a deterministic model with the shocks saved from
the estimation by specifying the source file for the shocks, using the
shocks(shocks file = datafile.mat) command. But of course, this is a bit
of a workaround, since you could also use the built-in commands in Dynare
to generate impulse response functions from estimated shocks, as described in
chapter 5.
4.2. FILES CREATED BY DYNARE 43
4.1.5 Infinite eigenvalues
If you use the command check in your .mod file, Dynare will report your sys-
tem’s eigenvalues and tell you if these meet the Blanchard-Kahn conditions.
At that point, don’t worry if you get infinite eigenvalues - these are are firmly
grounded in the theory of generalized eigenvalues. They have no detrimental
influence on the solution algorithm. As far as Blanchard-Kahn conditions are
concerned infinite eigenvalues are counted as explosive roots of modulus larger
than one.
4.2 Files created by Dynare
At times, you may get a message that there is an error in a file with a new
name, or you may want to have a closer look at how Dynare actually solves
your model - out of curiosity or maybe to do some customization of your own.
You may therefore find it helpful to get a brief overview of the internal files
that Dynare generates and the function of each one.
The dynare pre-processors essentially does three successive tasks: (i) pars-
ing of the mod file (it checks that the mod file is syntactically correct), and its
translation into internal machine representation (in particular, model equa-
tions are translated into expression trees), (ii) symbolic derivation of the model
equations, up to the needed order (depending on the computing needs), (iii)
outputting of several files, which are used from matlab. If the mod file is
“filename.mod”, then the pre-processor creates the following files:
• filename.m: a matlab file containing several instructions, notably the
parameter initializations and the matlab calls corresponding to comput-
ing tasks
• filename dynamic.m: a matlab file containing the model equations
and their derivatives (first, second and maybe third). Endogenous vari-
ables (resp. exogenous variables, parameters) are contained in a “y”
(resp. “x”, “params”) vector, with an index number depending on the
declaration order. The “y” vector has as many entries as their are (vari-
able, lag) pairs in the declared model. The model equations residuals
are stored in a vector named “residuals”. The model jacobian is put in
“g1” matrix. Second (resp. third) derivatives are in “g2” matrix (resp.
“g3”). If the “use dll” option has been specified in the model decla-
ration, the pre-processor will output a C file (with .c extension) rather
than a matlab file. It is then compiled to create a library (DLL) file. Us-
ing a compiled C file is supposed to give better computing performance
in model simulation/estimation.
44 CHAPTER 4. SOLVING DSGE MODELS - ADVANCED TOPICS
• filename static.m: a matlab file containing the stationarized version of
the model (i.e. where lagged variables are replaced by current variables),
with its jacobian. Used to compute the steady state. Same notations
than the dynamic file. Replaced by a C file when “use dll” option is
specified.
4.3 Modeling tips
4.3.1 Stationarizing your model
Models in Dynare must be stationary, such that you can linearize them around
a steady state and return to steady state after a shock. Thus, you must first
stationarize your model, then linearize it, either by hand, or by letting Dynare
do the work. You can then reconstruct ex-post the non-stationary simulated
variables after running impulse response functions.
For deterministic models, the trick is to use only stationary variables in
t + 1. More generally, if y
t
is I(1), you can always write y
t+1
as y
t
+ dy
t+1
,
where dy
t
= y
t
− y
t−1
. Of course, you need to know the value of dy
t
at the
final equilibrium.
Note that in a stationary model, it is expected that variables will eventually
go back to steady state after the initial shock. If you expect to see a growing
curve for a variable, you are thinking about a growth model. Because growth
models are nonstationary, it is easier to work with the stationarized version of
such models. Again, if you know the trend, you can always add it back after
the simulation of the stationary components of the variables.
4.3.2 Expectations taken in the past
For instance, to enter the term E
t−1
y
t
, define s
t
= E
t
[y
t+1
] and then use s(−1)
in your .mod file. Note that, because of Jensen’s inequality, you cannot do
this for terms that enter your equation in a non-linear fashion. If you do have
non-linear terms on which you want to take expectations in the past, you
would need to apply the above manipulation to the entire equation, as if y
t
were an equation, not just a variable.
4.3.3 Infinite sums
Dealing with infinite sums is tricky in general, and needs particular care when
working with Dynare. The trick is to use a recursive representation of the
sum. For example, suppose your model included:

j=0
β
j
x
t+j
= 0,
4.3. MODELING TIPS 45
Note that the above can also be written by using an auxiliary variable S
t
,
defined as:
S
t

j=0
β
j
x
t+j
,
which can also be written in the following recursive manner:
S
t

j=0
β
j
x
t+j
= x
t
+

j=1
β
j
x
t+j
= x
t

j=0
β
j
x
t+1+j
≡ x
t
+S
t+1
This formulation turns out to be useful in problems of the following form:

j=0
β
j
x
t+j
= p
t

j=0
γ
j
y
t+j
,
which can be written as a recursive system of the form:
S1
t
= x
t
+βS1
t+1
,
S2
t
= y
t
+γS2
t+1
,
S1 = p
t
S2.
This is particularly helpful, for instance, in a Calvo type setting, as
illustrated in the following brief example. The RBC model with monopolistic
competition introduced in chapter 3 involved flexible prices. The extension
with sticky prices, `a la Calvo for instance, is instead typical of the new Key-
nesian monetary literature, exemplified by papers such as Clarida, Gali, and
Gertler (1999).
The optimal price for a firm resetting its price in period t, given that it
will be able to reset its price only with probability 1 −θ each period, is
p

t
(i) = µ + (1 −βθ)

k=0
(βθ)
k
E
t
[mc
n
t+k
(i)]
where µ is the markup, β is a discount factor, i represents a firm of the contin-
uum between 0 and 1, and mc
t
is marginal cost as described in the example in
chapter 3. The trouble, of course, is how to input this infinite sum into
Dynare?
It turns out that the Calvo price setting implies that the aggregate price
follows the equation of motion p
t
= θp
t−1
+ (1 − θ)p

t
, thus implying the
following inflation relationship π
t
= (1 − θ)(p

t
− p
t−1
). Finally, we can also
rewrite the optimal price setting equation, after some algebraic manipulations,
as
p

t
−p
t−1
= (1 −βθ)

k=0
(βθ)
k
E
t
[ ´ mc
t+k
] +

k=0
(βθ)
k
E
t

t+k
]
46 CHAPTER 4. SOLVING DSGE MODELS - ADVANCED TOPICS
where ´ mc
t+k
= mc
t+k
+µ is the deviation of the marginal cost from its natural
rate, defined as the marginal cost when prices are perfectly flexible.
The trick now is to note that the above can be written recursively, by
writing the right hand side as the first term of the sum (with k = 0) plus the
remainder of the sum, which can be written as the left hand side term scrolled
forward one period and appropriately discounted. Mathematically, this yields:
p

t
−p
t−1
= (1 −βθ) ´ mc
t+k

t
+β θE
t
[p

t+1
−p
t
]
which has gotten rid of our infinite sum! That would be enough for Dynare,
but for convenience, we can go one step further and write the above as
π
t
= βE
t

t+1
] +λ´ mc
t
where λ ≡
(1−θ)(1−βθ)
θ
, which is the recognizable inflation equation in the new
Keynesian (or new Neoclassical) monetary literature.
4.3.4 Infinite sums with changing timing of expectations
When you are not able to write an infinite sum recursively, as the index of
the expectations changes with each element of the sum, as in the following
example, a different approach than the one mentioned above is necessary.
Suppose your model included the following sum:
y
t
=

j=0
E
t−j
x
t
where y
t
and x
t
are endogenous variables.
In Dynare, the best way to handle this is to write out the first k terms
explicitly and enter each one in Dynare, such as: E
t−1
x
t
+E
t−2
x
t
+. . .+E
t−k
x
t
.
Chapter 5
Estimating DSGE models -
basics
As in the chapter 3, this chapter is structured around an example. The goal
of this chapter is to lead you through the basic functionality in Dynare to
estimate models using Bayesian techniques, so that by the end of the chapter
you should have the capacity to estimate a model of your own. This chapter
is therefore very practically-oriented and abstracts from the underlying com-
putations that Dynare undertakes to estimate a model; that subject is instead
covered in some depth in chapter 8, while more advanced topics of practical
appeal are discussed in chapter 6.
5.1 Introducing an example
The example introduced in this chapter is particularly basic. This is for two
reasons. First, we did not want to introduce yet another example in this sec-
tion; there’s enough new material to keep you busy. Instead, we thought it
would be easiest to simply continue working with the example introduced in
chapter 3 with which you are probably already quite familiar. Second, the
goal of the example in this chapter is really to explain features in context,
but not necessarily to duplicate a “real life scenario” you would face when
writing a paper. Once you feel comfortable with the content of this chapter,
though, you can always move on to chapter 6 where you will find a full-fledged
replication of a recent academic paper, featuring a non-stationary model.
Recall from chapter 3 that we are dealing with an RBC model with mo-
nopolistic competition. Suppose we had data on business cycle variations of
output. Suppose also that we thought our little RBC model did a good job of
reproducing reality. We could then use Bayesian methods to estimate the pa-
rameters of the model: α, the capital share of output, β, the discount factor,
47
48 CHAPTER 5. ESTIMATING DSGE MODELS - BASICS
δ, the depreciation rate, ψ, the weight on leisure in the utility function, ρ, the
degree of persistence in productivity, and , the markup parameter. Note that
in Bayesian estimation, the condition for undertaking estimation is that there
be at least as many shocks as there are observables (a less stringent condition
than for maximum likelihood estimation). It may be that this does not allow
you to identify all your parameters - yielding posterior distributions identical
to prior distributions - but the Bayesian estimation procedure would still run
successfully. Let’s see how to go about doing this.
5.2 Declaring variables and parameters
To input the above model into Dynare for estimation purposes, we must first
declare the model’s variables in the preamble of the .mod file. This is done
exactly as described in chapter 3 on solving DSGE models. We thus begin the
.mod file with:
var y c k i l y l w r z;
varexo e;
parameters beta psi delta alpha rho epsilon;
5.3 Declaring the model
Suppose that the equation of motion of technology is a stationary AR(1) with
an autoregressive parameter, ρ, less than one. The model’s variables would
therefore be stationary and we can proceed without complications. The al-
ternative scenario with non-stationary variables is more complicated and
dealt with in chapter 6 (in the additional example). In the stationary case,
our model block would look exactly as in chater 3:
model;
(1/c) = beta*(1/c(+1))*(1+r(+1)-delta);
psi*c/(1-l) = w;
c+i = y;
y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha);
w = y*((epsilon-1)/epsilon)*(1-alpha)/l;
r = y*((epsilon-1)/epsilon)*alpha/k(-1);
i = k-(1-delta)*k(-1);
y l = y/l;
z = rho*z(-1)+e;
5.4. DECLARING OBSERVABLE VARIABLES 49
end;
5.4 Declaring observable variables
This should not come as a surprise. Dynare must know which variables are
observable for the estimation procedure. NOTE! These variables must be
available in the data file, as explained in section 5.7 below. For the moment,
we write:
varobs y;
5.5 Specifying the steady state
Before Dynare estimates a model, it first linearizes it around a steady state.
Thus, a steady state must exist for the model and although Dynare can calcu-
late it, we must give it a hand by declaring approximate values for the steady
state. This is just as explained in details and according to the same syntax
outlined in chapter 3, covering the initval, steady and check commands.
In fact, as this chapter uses the same model as that outlined in chapter 3, the
steady state block will look exactly the same.
TIP! During estimation, in finding the posterior mode, Dynare recalcu-
lates the steady state of the model at each iteration of the optimization rou-
tine (more on this later), based on the newest round of parameters available.
Thus, by providing approximate initial values and relying solely on the built-
in Dynare algorithm to find the steady state (a numerical procedure), you will
significantly slow down the computation of the posterior mode. Dynare will
end up spending 60 to 70% of the time recalculating steady states. It is much
more efficient to write an external Matlab steady state file and let Dynare
use that file to find the steady state of your model by algebraic procedure. For
more details on writing an external Matlab file to find your model’s steady
state, please refer to section 3.6.3 of chapter 3.
5.6 Declaring priors
Priors play an important role in Bayesian estimation and consequently de-
serve a central role in the specification of the .mod file. Priors, in Bayesian
estimation, are declared as a distribution. The general syntax to introduce
priors in Dynare is the following:
50 CHAPTER 5. ESTIMATING DSGE MODELS - BASICS
estimated params;
PARAMETER NAME, PRIOR SHAPE, PRIOR MEAN, PRIOR STANDARD ERROR [, PRIOR
3
rd
PARAMETER] [,PRIOR 4
th
PARAMETER] ;
end;
where the following table defines each term more clearly
PRIOR SHAPE Corresponding distribution Range
NORMAL PDF N(µ, σ) R
GAMMA PDF G
2
(µ, σ, p
3
) [p
3
, +∞)
BETA PDF B(µ, σ, p
3
, p
4
) [p
3
, p
4
]
INV GAMMA PDF IG
1
(µ, σ) R
+
UNIFORM PDF U(p
3
, p
4
) [p
3
, p
4
]
where µ is the PRIOR MEAN, σ is the PRIOR STANDARD ERROR, p
3
is the PRIOR
3
rd
PARAMETER (whose default is 0) and p
4
is the PRIOR 4
th
PARAMETER (whose
default is 1). TIP! When specifying a uniform distribution between 0 and 1 as
a prior for a parameter, say α, you therefore have to put two empty spaces for
parameters µ and σ, and then specify parameters p3 and p4, since the uniform
distribution only takes p3 and p4 as arguments. For instance, you would write
alpha, uniform pdf, , , 0,1;
For a more complete review of all possible options for declaring priors, as
well as the syntax to declare priors for maximum likelihood estimation (not
Bayesian), see the Reference Manual. Note also that if some parameters in a
model are calibrated and are not to be estimated, you should declare them as
such, by using the parameters command and its related syntax, as explained
in chapter 3.
TIP! Choosing the appropriate prior for your parameters is a tricky, yet
very important endeavor. It is worth spending time on your choice of priors
and to test the robustness of your results to your priors. Some considerations
may prove helpful. First, think about the domain of your prior over each pa-
rameter. Should it be bounded? Should it be opened on either or both sides?
Remember also that if you specify a probability of zero over a certain domain
in your prior, you will necessarily also find a probability of zero in your pos-
terior distribution. Then, think about the shape of your prior distribution.
Should it be symmetric? Skewed? If so, on which side? You may also go
one step further and build a distribution for each of your parameters in your
mind. Ask yourself, for instance, what is the probability that your parameter
is bigger than a certain value, and repeat the exercise by incrementally de-
creasing that value. You can then pick the standard distribution that best fits
5.6. DECLARING PRIORS 51
your perceived distribution. Finally, instead of describing here the shapes and
properties of each standard distribution available in Dynare, you are instead
encouraged to visualize these distributions yourself, either in a statistics book
or on the Web.
TIP! It may at times be desirable to estimate a transformation of a pa-
rameter appearing in the model, rather than the parameter itself. In such a
case, it is possible to declare the parameter to be estimated in the parameters
statement and to define the transformation at the top of the model section,
as a Matlab expression, by adding a pound sign (#) at the beginning of the
corresponding line. For example, you may find it easier to define a prior over
the discount factor, β, than its inverse which often shows up in Euler equa-
tions. Thus you would write:
model;
# sig = 1/bet;
c = sig*c(+1)*mpk;
end;
estimated params;
bet,normal pdf,1,0.05;
end;
TIP! Finally, another useful command to use is the estimated params init
command which declares numerical initial values for the optimizer when these
are different from the prior mean. This is especially useful when redoing an
estimation - if the optimizer got stuck the first time around, or needing a
greater number of draws in the Metropolis-Hastings algorithm - and wanting
to enter the posterior mode as initial values for the parameters instead of a
prior. The Reference Manual gives more details as to the exact syntax of this
command.
Coming back to our basic example, we would write:
estimated params;
alpha, beta pdf, 0.35, 0.02;
beta, beta pdf, 0.99, 0.002;
delta, beta pdf, 0.025, 0.003;
psi, gamma pdf, 1.75, 0.02;
rho, beta pdf, 0.95, 0.05;
epsilon, gamma pdf, 10, 0.003;
stderr e, inv gamma pdf, 0.01, inf;
end;
52 CHAPTER 5. ESTIMATING DSGE MODELS - BASICS
5.7 Launching the estimation
To ask Dynare to estimate a model, all that is necessary is to add the com-
mand estimation at the end of the .mod file. Easy enough. But the real
complexity comes from the options available for the command (to be entered
in parentheses and sequentially, separated by commas, after the command
estimation). Below, we list the most common and useful options, and en-
courage you to view the Reference Manual for a complete list.
1. datafile = FILENAME: the datafile (a .m file, a .mat file, or an .xls
file). Note that observations do not need to show up in any order, but
vectors of observations need to be named with the same names as those
in var obs. In Excel files, for instance, observations could be ordered
in columns, and variable names would show up in the first cell of each
column.
2. nobs = INTEGER: the number of observations to be used (default: all
observations in the file)
3. first obs = INTEGER: the number of the first observation to be used
(default = 1). This is useful when running loops, or instance, to divide
the observations into sub-periods.
4. prefilter = 1: the estimation procedure demeans the data (default=0,
no prefiltering). This is useful if model variables are in deviations from
steady state, for instance, and therefore have zero mean. Demeaning the
observations would also impose a zero mean on the observed variables.
5. nograph: no graphs should be plotted.
6. conf sig = {INTEGER — DOUBLE }: the level for the confidence in-
tervals reported in the results (default = 0.90)
7. mh replic = INTEGER: number of replication for Metropolis Hasting
algorithm. For the time being, mh replic should be larger than 1200
(default = 20000)
8. mh nblocks = INTEGER: number of parallel chains for Metropolis Hast-
ing algorithm (default = 2). Despite this low default value, it is advisable
to work with a higher value, such as 5 or more. This improves the com-
putation of between group variance of the parameter means, one of the
key criteria to evaluate the efficiency of the Metropolis-Hastings to eval-
uate the posterior distribution. More details on this subject appear in
Chapter 6.
5.7. LAUNCHING THE ESTIMATION 53
9. mh drop = DOUBLE: the fraction of initially generated parameter vec-
tors to be dropped before using posterior simulations (default = 0.5; this
means that the first half of the draws from the Metropolis-Hastings are
discarded).
10. mh jscale = DOUBLE: the scale to be used for the jumping distribu-
tion in MH algorithm. The default value is rarely satisfactory. This
option must be tuned to obtain, ideally, an acceptance rate of 25% in
the Metropolis- Hastings algorithm (default = 0.2). The idea is not to
reject or accept too often a candidate parameter; the literature has set-
tled on a value of between 0.2 and 0.4. If the acceptance rate were too
high, your Metropolis-Hastings iterations would never visit the tails of a
distribution, while if it were too low, the iterations would get stuck in a
subspace of the parameter range. Note that the acceptance rate drops if
you increase the scale used in the jumping distribution and vice a versa.
11. mh init scale=DOUBLE: the scale to be used for drawing the initial
value of the Metropolis-Hastings chain (default=2*mh jscale). The idea
here is to draw initial values from a stretched out distribution in order
to maximize the chances of these values not being too close together,
which would defeat the purpose of running several blocks of Metropolis-
Hastings chains.
12. mode file=FILENAME: name of the file containing previous value for
the mode. When computing the mode, Dynare stores the mode (xparam1)
and the hessian (hh) in a file called MODEL NAME mode. This is a
particularly helpful option to speed up the estimation process if you have
already undertaken initial estimations and have values of the posterior
mode.
13. mode compute=INTEGER: specifies the optimizer for the mode com-
putation.
0: the mode isn’t computed. mode file must be specified
1: uses Matlab fmincon (see the Reference Manual to set options
for this command).
2: uses Lester Ingber’s Adaptive Simulated Annealing.
3: uses Matlab fminunc.
4 (default): uses Chris Sim’s csminwel.
14. mode check: when mode check is set, Dynare plots the minus of the
posterior density for values around the computed mode for each esti-
mated parameter in turn. This is helpful to diagnose problems with the
optimizer. A clear indication of a problem would be that the mode is
not at the trough (bottom of the minus) of the posterior distribution.
54 CHAPTER 5. ESTIMATING DSGE MODELS - BASICS
15. load mh file: when load mh file is declared, Dynare adds to previous
Metropolis-Hastings simulations instead of starting from scratch. Again,
this is a useful option to speed up the process of estimation.
16. nodiagnostic: doesn’t compute the convergence diagnostics for Metropolis-
Hastings (default: diagnostics are computed and displayed). Actually
seeing if the various blocks of Metropolis-Hastings runs converge is a
powerful and useful option to build confidence in your model estima-
tion. More details on these diagnostics are given in Chapter 6.
17. bayesian irf: triggers the computation of the posterior distribution of im-
pulse response functions (IRFs). The length of the IRFs are controlled
by the irf option, as specified in chapter 3 when discussing the options for
stoch simul. To build the posterior distribution of the IRFs, Dynare
pulls parameter and shock values from the corresponding estimated dis-
tributions and, for each set of draws, generates an IRF. Repeating this
process often enough generates a distribution of IRFs. TIP! If you stop
the estimation procedure after calculating the posterior mode, or carry
out maximum likelihood estimation, only the corresponding parameter
estimates will be used to generate the IRFs. If you instead carry out
a full Metropolis-Hastings estimation, on the other hand, the IRFs will
use the parameters the posterior distributions, including the variance of
the shocks.
18. All options available for stoch simul can simply be added to the above
options, separated by commas. To view a list of these options, either
see the Reference Manual or section 3.8 of chapter 3.
19. moments varendo: triggers the computation of the posterior distribution
of the theoretical moments of the endogenous variables as in stoch simul
(the posterior distribution of the variance decomposition is also in-
cluded). ** will be implemented shortly - if not already - in Dynare
version 4.
20. filtered vars: triggers the computation of the posterior distribution of
filtered endogenous variables and shocks. See the note below on the
difference between filtered and smoothed shocks. ** will be implemented
shortly - if not already - in Dynare version 4.
21. smoother: triggers the computation of the posterior distribution of smoothed
endogenous variables and shocks. Smoothed shocks are a reconstruction
of the values of unobserved shocks over the sample, using all the informa-
tion contained in the sample of observations. Filtered shocks, instead,
are built only based on knowing past information. To calculate one pe-
riod ahead prediction errors, for instance, you should use filtered, not
smoothed variables.
5.8. THE COMPLETE .MOD FILE 55
22. forecast = INTEGER: computes the posterior distribution of a forecast
on INTEGER periods after the end of the sample used in estimation.
The corresponding graph includes one confidence interval describing un-
certainty due to parameters and one confidence interval describing un-
certainty due to parameters and future shocks. Note that Dynare cannot
forecast out of the posterior mode. You need to run Metropolis-Hastings
iterations before being able to run forecasts on an estimated model. Fi-
nally, running a forecast is very similar to an IRF, as in bayesian irf,
except that the forecast does not begin at a steady state, but simply
at the point corresponding to the last set of observations. The goal
of undertaking a forecast is to see how the system returns to steady
state from this starting point. Of course, as observation do not exist
for all variables, those necessary are reconstructed by sampling out of
the posterior distribution of parameters. Again, repeating this step of-
ten enough yields a posterior distribution of the forecast. ** will be
implemented shortly - if not already - in Dynare version 4.
TIP! Before launching estimation it is a good idea to make sure that your
model is correctly declared, that a steady state exists and that it can be sim-
ulated for at least one set of parameter values. You may therefore want to
create a test version of your .mod file. In this test file, you would comment
out or erase the commands related to estimation, remove the prior estimates
for parameter values and replace them with actual parameter values in the
preamble, remove any non-stationary variables from your model, add a shocks
block, make sure you have steady and possibly check following the initval
block if you do not have exact steady state values and run a simulation using
stoch simul at the end of your .mod file. Details on model solution and sim-
ulation can be found in Chapter 3.
Finally, coming back to our example, we could choose a standard option:
estimation(datafile=simuldataRBC,nobs=200,first obs=500,
mh replic=2000,mh nblocks=2,mh drop=0.45,mh jscale=0.8,
mode compute=6);
This ends our description of the .mod file.
5.8 The complete .mod file
To summarize and to get a complete perspective on our work so far, here is
the complete .mod file for the estimation of our very basic model. You can
find the corresponding file in the models folder under UserGuide in your in-
stallation of Dynare. The file is called RBC Est.mod.
56 CHAPTER 5. ESTIMATING DSGE MODELS - BASICS
var y c k i l y l w r z;
varexo e;
parameters beta psi delta alpha rho epsilon;
model;
(1/c) = beta*(1/c(+1))*(1+r(+1)-delta);
psi*c/(1-l) = w;
c+i = y;
y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha);
w = y*((epsilon-1)/epsilon)*(1-alpha)/l;
r = y*((epsilon-1)/epsilon)*alpha/k(-1);
i = k-(1-delta)*k(-1);
y l = y/l;
z = rho*z(-1)+e;
end;
varobs y;
initval;
k = 9;
c = 0.7;
l = 0.3;
w = 2.0;
r = 0;
z = 0;
e = 0;
end;
steady;
check;
estimated params;
alpha, beta pdf, 0.35, 0.02;
beta, beta pdf, 0.99, 0.002;
delta, beta pdf, 0.025, 0.003;
psi, gamma pdf, 1.75, 0.02;
rho, beta pdf, 0.95, 0.05;
epsilon, gamma pdf, 10, 0.003;
stderr e, inv gamma pdf, 0.01, inf;
end;
estimation(datafile=simuldataRBC,nobs=200,first obs=500,
mh replic=2000,mh nblocks=2,mh drop=0.45,mh jscale=0.8,
mode compute=6);
5.9. INTERPRETING OUTPUT 57
5.9 Interpreting output
As in the case of model solution and simulation, Dynare returns both tabular
and graphical output. On the basis of the options entered in the example
.mod file above, Dynare will display the following results.
5.9.1 Tabular results
The first results to be displayed (and calculated from a chronological stand-
point) are the steady state results. Note the dummy values of 1 for the non-
stationary variables Y obs and P obs. These results are followed by the eigen-
values of the system, presented in the order in which the endogenous variables
are declared at the beginning of the .mod file. The table of eigenvalues is
completed with a statement about the Blanchard-Kahn condition being met
- hopefully!
The next set of results are for the numerical iterations necessary to find
the posterior mode, as explained in more details in Chapter 6. The improve-
ment from one iteration to the next reaches zero, Dynare give the value of
the objective function (the posterior Kernel) at the mode and displays two
important table summarizing results from posterior maximization.
The first table summarizes results for parameter values. It includes: prior
means, posterior mode, standard deviation and t-stat of the mode (based on
the assumption of a Normal, probably erroneous when undertaking Bayesian
estimation, as opposed to standard maximum likelihood), as well as the prior
distribution and standard deviation (pstdev). It is followed by a second table
summarizing the same results for the shocks. It may be entirely possible that
you get an infinite value for a standard deviation, this is simply the limit case
of the inverse gamma distribution.
5.9.2 Graphical results
** corresponding graphs will be reproduced below.
The first figure comes up soon after launching Dynare as little computa-
tion is necessary to generate it. The figure returns a graphical representation
of the priors for each parameter of interest.
58 CHAPTER 5. ESTIMATING DSGE MODELS - BASICS
The second set of figures displays “MCMC univariate diagnostics”, where
MCMC stands for Monte Carlo Markov Chains. This is the main source
of feedback to gain confidence, or spot a problem, with results. Recall that
Dynare completes several runs of Metropolis-Hastings simulations (as many as
determined in the option mh nblocks, each time starting from a different ini-
tial value). If the results from one chain are sensible, and the optimizer did not
get stuck in an odd area of the parameter subspace, two things should happen.
First, results within any of the however many iterations of Metropolis-Hastings
simulation should be similar. And second, results between the various chains
should be close. This is the idea of what the MCMC diagnostics track.
More specifically, the red and blue lines on the charts represent specific
measures of the parameter vectors both within and between chains. For the
results to be sensible, these should be relatively constant (although there
will always be some variation) and they should converge. Dynare reports
three measures: “interval”, being constructed from an 80% confidence inter-
val around the parameter mean, “m2”, being a measure of the variance and
“m3” based on third moments. In each case, Dynare reports both the within
and the between chains measures. The figure entitled “multivariate diagnos-
tic” presents results of the same nature, except that they reflect an aggregate
measure based on the eigenvalues of the variance-covariance matrix of each
parameter.
In our example above, you can tell that indeed, we obtain convergence
and relative stability in all measures of the parameter moments. Note that
the horizontal axis represents the number of Metropolis-Hastings iterations
that have been undertaken, and the vertical axis the measure of the parame-
ter moments, with the first, corresponding to the measure at the initial value
of the Metropolis-Hastings iterations.
TIP! If the plotted moments are highly unstable or do not converge, you
may have a problem of poor priors. It is advisable to redo the estimation with
different priors. If you have trouble coming up with a new prior, try starting
with a uniform and relatively wide prior and see where the data leads the
posterior distribution. Another approach is to undertake a greater number of
Metropolis-Hastings simulations.
The first to last figure - figure 6 in our example - displays the most inter-
esting set of results, towards which most of the computations undertaken by
Dynare are directed: the posterior distribution. In fact, the figure compares
the posterior to the prior distribution (black vs. grey lines). In addition,
on the posterior distribution, Dynare plots a green line which represents the
posterior mode. These allow you to make statements about your data other
than simply concerning the mean and variance of the parameters; you can also
5.9. INTERPRETING OUTPUT 59
discuss the probability that your parameter is larger or smaller than a certain
value.
TIP! These graphs are of course especially relevant and present key results,
but they can also serve as tools to detect problems or build additional confi-
dence in your results. First, the prior and the posterior distributions should
not be excessively different. Second, the posterior distributions should be close
to normal, or at least not display a shape that is clearly non-normal. Third,
the green mode (calculated from the numerical optimization of the posterior
kernel) should not be too far away from the mode of the posterior distribution.
If not, it is advisable to undertake a greater number of Metropolis-Hastings
simulations.
The last figure returns the smoothed estimated shocks in a useful illustra-
tion to eye-ball the plausibility of the size and frequency of the shocks. The
horizontal axis, in this case, represents the number of periods in the sample.
One thing to check is the fact that shocks should be centered around zero.
That is indeed the case for our example.
Chapter 6
Estimating DSGE models -
advanced topics
This chapter focusses on advanced topics and features of Dynare in the area of
model estimation. The chapter begins by presenting a more complex example
than the one used for illustration purposes in chapter 5. The goal is to show
how Dynare would be used in the more “realistic” setting of reproducing a
recent academic paper. The chapter then follows with sections on comparing
models to one another, and then to BVARs, and ends with a table summariz-
ing where output series are stored and how these can be retrieved.
6.1 Alternative and non-stationary example
The example provided in chapter 5 is really only useful for illustration pur-
poses. So we thought you would enjoy (and continue learning from!) a more
realistic example which reproduces the work in a recent - and highly regarded
- academic paper. The example shows how to use Dynare in a more realistic
setting, while emphasizing techniques to deal with non-stationary observations
and stochastic trends in dynamics.
6.1.1 Introducing the example
The example is drawn from Schorfheide (2000). This first section introduces
the model, its basic intuitions and equations. We will then see in subsequent
sections how to estimate it using Dynare. Note that the original paper by
Schorfheide mainly focusses on estimation methodologies, difficulties and so-
lutions, with a special interest in model comparison, while the mathematics
and economic intuitions of the model it evaluates are drawn from Nason and
Cogley (1994). That paper should serve as a helpful reference if anything is
61
62 CHAPTER 6. ESTIMATING DSGE MODELS - ADVANCED TOPICS
Figure 6.1: Continuous lines show the circulation of nominal funds, while
dashed lines show the flow of real variables.
left unclear in the description below.
In essence, the model studied by Schorfheide (2000) is one of cash in ad-
vance (CIA). The goal of the paper is to estimate the model using Bayesian
techniques, while observing only output and inflation. In the model, there are
several markets and actors to keep track of. So to clarify things, figure 6.1.1
sketches the main dynamics of the model. You may want to refer back to the
figure as you read through the following sections.
The economy is made up of three central agents and one secondary agent:
households, firms and banks (representing the financial sector), and a mon-
etary authority which plays a minor role. Households maximize their utility
function which depends on consumption, C
t
, and hours worked, H
t
, while
deciding how much money to hold next period in cash, M
t+1
and how much
to deposit at the bank, D
t
, in order to earn R
H,t
− 1 interest. Households
6.1. ALTERNATIVE AND NON-STATIONARY EXAMPLE 63
therefore solve the problem
max
{C
t
,H
t
,M
t+1
,D
t
}
E
0
_

t=0
β
t
[(1 −φ) ln C
t
+φln(1 −H
t
)]
¸
s.t. P
t
C
t
≤ M
t
−D
t
+W
t
H
t
0 ≤ D
t
M
t+1
= (M
t
−D
t
+W
t
H
t
−P
t
C
t
) +R
H,t
D
t
+F
t
+B
t
where the second equation spells out the cash in advance constraint including
wage revenues, the third the inability to borrow from the bank and the fourth
the intertemporal budget constraint emphasizing that households accumulate
the money that remains after bank deposits and purchases on goods are de-
ducted from total inflows made up of the money they receive from last period’s
cash balances, wages, interests, as well as dividends from firms, F
t
, and from
banks, B
t
, which in both cases are made up of net cash inflows defined below.
Banks, on their end, receive cash deposits from households and a cash
injection, X
t
from the central bank (which equals the net change in nominal
money balances, M
t+1
− M
t
). It uses these funds to disburse loans to firms,
L
t
, on which they make a net return of R
F,t
− 1. Of course, banks are con-
strained in their loans by a credit market equilibrium condition L
t
≤ X
t
+D
t
.
Finally, bank dividends, B
t
are simply equal to D
t
+R
F,t
L
t
−R
H,t
D
t
−L
t
+X
t
.
Finally, firms maximize the net present value of future dividends (dis-
counted by the marginal utility of consumption, since they are owned by
households) by choosing dividends, next period’s capital stock, K
t+1
, labor
demand, N
t
, and loans. Its problem is summarized by
max
{F
t
,K
t+1
,N
t
,L
t
}
E
0
_


t=0
β
t+1 F
t
C
t+1
P
t+1
_
s.t. F
t
≤ L
t
+P
t
_
K
α
t
(A
t
N
t
)
1−α
−K
t+1
+ (1 −δ)K
t
¸
−W
t
N
t
−L
t
R
F,t
W
t
N
t
≤ L
t
where the second equation makes use of the production function Y
t
= K
α
t
(A
t
N
t
)
1−α
and the real aggregate accounting constraint (goods market equilibrium) C
t
+I
t
= Y
t
,
where I
t
= K
t+1
− (1 − δ)K
t
, and where δ is the rate of depreciation. Note
that it is the firms that engage in investment in this model, by trading off
investment for dividends to consumers. The third equation simply specifies
that bank loans are used to pay for wage costs.
To close the model, we add the usual labor and money market equilib-
rium equations, H
t
= N
t
and P
t
C
t
= M
t
+ X
t
, as well as the condition that
R
H,t
= R
F,t
due to the equal risk profiles of the loans.
64 CHAPTER 6. ESTIMATING DSGE MODELS - ADVANCED TOPICS
More importantly, we add a stochastic elements to the model. The model
allows for two sources of perturbations, one real, affecting technology and one
nominal, affecting the money stock. These important equations are
ln A
t
= γ + ln A
t−1
+
A,t
,
A,t
∼ N(0, σ
2
A
)
and
ln m
t
= (1 −ρ) ln m

+ρ ln m
t−1
+
M,t
,
M,t
∼ N(0, σ
2
M
)
where m
t
≡ M
T+1
/M
t
is the growth rate of the money stock. Note that theses
expressions for trends are not written in the most straightforward manner nor
very consistently. But we reproduced them never-the-less to make it easier to
compare this example to the original paper.
The first equation is therefore a unit root with drift in the log of tech-
nology, and the second an autoregressive stationary process in the growth
rate of money, but an AR(2) with a unit root in the log of the level of
money. This can be seen from the definition of m
t
which can be rewritten
as ln M
t+1
= ln M
t
+ ln m
t
.
1
When the above functions are maximized, we obtain the following set of
first order and equilibrium conditions. We will not dwell on the derivations
here, to save space, but encourage you to browse Nason and Cogley (1994) for
additional details. We nonetheless give a brief intuitive explanation of each
1
Alternatively, we could have written the AR(2) process in state space form and realized
that the system has an eigenvalue of one. Otherwise said, one is a root of the second order
autoregressive lag polynomial. As usual, if the logs of a variable are specified to follow a unit
root process, the rate of growth of the series is a stationary stochastic process; see Hamilton
(1994), chapter 15, for details.
6.1. ALTERNATIVE AND NON-STATIONARY EXAMPLE 65
equation. The system comes down to
E
t
_

´
P
t
/
_
´
C
t+1
´
P
t+1
m
t
_
_
= βe
−α(γ+
A,t+1
)
P
t+1
_
α
´
K
α−1
t
N
1−α
t+1
+ (1 −δ)
_
/
_
´c
t+2
´
P
t+2
m
t+1
_
_
´
W
t
=
´
L
t
/N
t
φ
1 −φ
_
´
C
t
´
P
t
/ (1 −N
t
)
_
=
´
L
t
/N
t
R
t
= (1 −α)
´
P
t
e
−α(γ+
A,t+1
)
´
K
α
t−1
N
−α
t
/
´
W
t
_
´
C
t
´
P
t
_
−1
= β
_
(1 −α)
´
P
t
e
−α(γ+
A,t+1
)
´
K
α
t−1
N
1−α
t
_
×E
t
_
´
L
t
m
t
´
C
t+1
´
P
t+1
_
−1
´
C
t
+
´
K
t
= e
−α(γ+
A,t
)
´
K
α
t−1
N
1−α
+ (1 −δ)e
−(γ+
A,t
)
´
K
t−1
´
P
t
´
C = m
t
m
t
−1 +
´
D
t
=
´
L
t
´
Y
t
=
´
K
α
t−1
N
1−α
e
−α(γ+
A,t
)
ln(m
t
) = (1 −ρ) ln(m

) +ρ ln(m
t−1
) +
M,t
A
t
A
t−1
≡ dA
t
= exp(γ +
A,t
)
Y
t
/Y
t−1
= e
γ+
A,t ´
Y
t
/
´
Y
t−1
P
t
/P
t−1
= (
´
P
t
/
´
P
t−1
)(m
t−1
/e
γ+
A,t
)
where, importantly, hats over variables no longer mean deviations from steady
state, but instead represent variables that have been made stationary. We
come back to this important topic in details in section 6.1.3 below. For now,
we pause a moment to give some intuition for the above equations. In order,
these equations correspond to:
1. The Euler equation in the goods market, representing the tradeoff to the
economy of moving consumption goods across time.
2. The firms’ borrowing constraint, also affecting labor demand, as firms
use borrowed funds to pay for labor input.
3. The intertemporal labor market optimality condition, linking labor sup-
ply, labor demand, and the marginal rate of substitution between con-
sumption and leisure.
4. The equilibrium interest rate in which the marginal revenue product of
labor equals the cost of borrowing to pay for that additional unit of
labor.
66 CHAPTER 6. ESTIMATING DSGE MODELS - ADVANCED TOPICS
5. The Euler equation in the credit market, which ensures that giving up
one unit of consumption today for additional savings equals the net
present value of future consumption.
6. The aggregate resource constraint.
7. The money market equilibrium condition equating nominal consumption
demand to money demand to money supply to current nominal balances
plus money injection.
8. The credit market equilibrium condition.
9. The production function.
10. The stochastic process for money growth.
11. The stochastic process for technology.
12. The relationship between observable variables and stationary variables;
more details on these last two equations appear in the following section.
6.1.2 Declaring variables and parameters
This block of the .mod file follows the usual conventions and would look like:
var m P c e W R k d n l Y obs P obs y dA;
varexo e a e m;
parameters alp, bet, gam, mst, rho, psi, del;
where the choice of upper and lower case letters is not significant, the first set
of endogenous variables, up to l, are as specified in the model setup above,
and where the last five variables are defined and explained in more details in
the section below on declaring the model in Dynare. The exogenous variables
are as expected and concern the shocks to the evolution of technology and
money balances.
6.1.3 The origin of non-stationarity
The problem of non-stationarity comes from having stochastic trends in tech-
nology and money. The non-stationarity comes out clearly when attempting
to solve the model for a steady state and realizing it does not have one. It can
be shown that when shocks are null, real variables grow with A
t
(except for
labor, N
t
, which is stationary as there is no population growth), nominal vari-
ables grow with M
t
and prices with M
t
/A
t
. Detrending therefore involves
6.1. ALTERNATIVE AND NON-STATIONARY EXAMPLE 67
the following operations (where hats over variables represent stationary vari-
ables): for real variables, ˆ q
t
= q
t
/A
t
, where q
t
= [y
t
, c
t
, i
t
, k
t+1
]. For nominal
variables,
ˆ
Q
t
= Q
t
/M
t
, where Q
t
= [d
t
, l
t
, W
t
]. And for prices,
ˆ
P
t
= P
t
·A
t
/M
t
.
6.1.4 Stationarizing variables
Let’s illustrate this transformation on output, and leave the transformations
of the remaining equations as an exercise, if you wish (Nason and Cogley
(1994) includes more details on the transformations of each equation). We
stationarize output by dividing its real variables (except for labor) by A
t
. We
define
´
Y
t
to equal Y
t
/A
t
and
´
K
t
as K
t
/A
t
. NOTE! Recall from section 3.5 in
chapter 3), that in Dynare variables take the time subscript of the period in
which they are decided (in the case of the capital stock, today’s capital stock
is a result of yesterday’s decision). Thus, in the output equation, we should
actually work with
´
K
t−1
= K
t−1
/A
t−1
. The resulting equation made up of
stationary variables is
Y
t
A
t
=
_
K
t−1
A
t−1
_
α
A
1−α
t
N
1−α
t
A
−1
t
A
α
t−1
´
Y
t
=
´
K
α
t−1
N
1−α
t
_
A
t
A
t−1
_
−α
=
´
K
α
t−1
N
1−α
t
exp(−α(γ +
A,t
))
where we go from the second to the third line by taking the exponential of
both sides of the equation of motion of technology.
The above is the equation we retain for the .mod file of Dynare into which
we enter:
y=k(-1)^alp*n^(1-alp)*exp(-alp*(gam+e a))
The other equations are entered into the .mod file after transforming them
in exactly the same way as the one above. A final transformation to consider,
that turns out to be useful since we often deal with the growth rate of tech-
nology, is to define
dA = exp(gam+e a)
by simply taking the exponential of both sides of the stochastic process of
technology defined in the model setup above.
68 CHAPTER 6. ESTIMATING DSGE MODELS - ADVANCED TOPICS
6.1.5 Linking stationary variables to the data
And finally, we must make a decision as to our non-stationary observa-
tions. We could simply stationarize them by working with rates of growth
(which we know are constant). In the case of output, the observable variable
would become Y
t
/Y
t−1
. We would then have to relate this observable, call it
gy obs, to our (stationary) model’s variables
´
Y
t
by using the definition that
´
Y
t
≡ Y
t
/A
t
. Thus, we add to the model block of the .mod file:
gy obs = dA*y/y(-1);
where, the y of the .mod file are the stationary
´
Y
t
.
But, we could also work with non-stationary data in levels. This
complicates things somewhat, but illustrates several features of Dynare worth
highlighting; we therefore follow this path in the remainder of the example.
The result is not very different, though, from what we just saw above. The
goal is to add a line to the model block of our .mod file that relates the non
stationary observables, call them Y
obs
, to our stationary output,
´
Y
t
. We could
simply write Y
obs
=
´
Y
t
A
t
. But since we don’t have an A
t
variable, but just a
dA
t
, we we-write the above relationship in ratios. To the .mod file, we there-
fore add:
Y obs/Y obs(-1) = dA*y/y(-1);
We of course do the same for prices, our other observable variable, except
that we use the relationship P
obs
=
´
P
t
M
t
/A
t
as noted earlier. The details
of the correct transformations for prices are left as an exercise and can be
checked against the results below.
6.1.6 The resulting model block of the .mod file
model;
dA = exp(gam+e a);
log(m) = (1-rho)*log(mst) + rho*log(m(-1))+e m;
-P/(c(+1)*P(+1)*m)+bet*P(+1)*(alp*exp(-alp*(gam+log(e(+1))))*k^(alp-1)
*n(+1)^(1-alp)+(1-del)*exp(-(gam+log(e(+1)))))/(c(+2)*P(+2)*m(+1))=0;
W = l/n;
-(psi/(1-psi))*(c*P/(1-n))+l/n = 0;
R = P*(1-alp)*exp(-alp*(gam+e a))*k(-1)^alp*n^(-alp)/W;
1/(c*P)-bet*P*(1-alp)*exp(-alp*(gam+e a))*k(-1)^alp*n^(1-alp)/
(m*l*c(+1)*P(+1)) = 0;
c+k = exp(-alp*(gam+e a))*k(-1)^alp*n^(1-alp)+(1-del)
6.1. ALTERNATIVE AND NON-STATIONARY EXAMPLE 69
*exp(-(gam+e a))*k(-1);
P*c = m;
m-1+d = l;
e = exp(e a);
y = k(-1)^alp*n^(1-alp)*exp(-alp*(gam+e a));
Y obs/Y obs(-1) = dA*y/y(-1);
P obs/P obs(-1) = (p/p(-1))*m(-1)/dA;
end;
where, of course, the input conventions, such as ending lines with semicolons
and indicating the timing of variables in parentheses, are the same as those
listed in chapter 3.
TIP! In the above model block, notice that what we have done is in fact
relegated the non-stationarity of the model to just the last two equations,
concerning the observables which are, after all, non-stationary. The problem
that arises, though, is that we cannot linearize the above system in levels, as
the last two equations don’t have a steady state. If we first take logs, though,
they become linear and it doesn’t matter anymore where we calculate their
derivative when taking a Taylor expansion of all the equations in the system.
Thus, when dealing with non-stationary observations, you must log-
linearize your model (and not just linearize it); this is a point to which we
will return later.
6.1.7 Declaring observable variables
We begin by declaring which of our model’s variables are observables. In our
.mod file we write
varobs P obs Y obs;
to specify that our observable variables are indeed P obs and Y obs as noted
in the section above. NOTE! Recall from earlier that the number of observed
variables must be smaller or equal to the number of shocks such that the model
be estimated. If this is not the case, you should add measurement shocks to
your model where you deem most appropriate.
6.1.8 Declaring trends in observable variables
Recall that we decided to work with the non-stationary observable variables
in levels. Both output and prices exhibit stochastic trends. This can be seen
explicitly by taking the difference of logs of output and prices to compute
growth rates. In the case of output, we make use of the usual (by now!)
70 CHAPTER 6. ESTIMATING DSGE MODELS - ADVANCED TOPICS
relationship Y
t
=
´
Y
t
· A
t
. Taking logs of both sides and subtracting the same
equation scrolled back one period, we find:
∆ln Y
t
= ∆ln
´
Y
t
+γ +
A,t
emphasizing clearly the drift term γ, whereas we know ∆ln
´
Y
t
is stationary in
steady state.
In the case of prices, we apply the same manipulations to show that:
∆ln P
t
= ∆ln
´
P
t
+ ln m
t−1
−γ −
A,t
Note from the original equation of motion of ln m
t
that in steady state,
ln m
t
= ln m

, so that the drift terms in the above equation are ln m

−γ.
2
In Dynare, any trends, whether deterministic or stochastic (the drift term)
must be declared up front. In the case of our example, we therefore write (in
a somewhat cumbersome manner)
observation trends;
P obs (log(mst)-gam);
Y obs (gam);
end;
In general, the command observation trends specifies linear trends as a
function of model parameters for the observed variables in the model.
6.1.9 Declaring unit roots in observable variables
And finally, since P obs and Y obs inherit the unit root characteristics of their
driving variables, technology and money, we must tell Dynare to use a diffuse
prior (infinite variance) for their initialization in the Kalman filter. Note that
for stationary variables, the unconditional covariance matrix of these variables
is used for initialization. The algorithm to compute a true diffuse prior is taken
from Durbin and Koopman (2001). To give these instructions to Dynare, we
write in the .mod
unit root vars P obs Y obs;
NOTE! You don’t need to declare unit roots for any non-stationary model.
Unit roots are only related to stochastic trends. You don’t need to use a diffuse
2
This can also be see from substituting for ln m
t−1
in the above equation with the
equation of motion of ln m
t
to yield: ∆ln P
t
= ∆ln

P
t
+ln m

+ρ(ln m
t−2
−ln m

) +
M,t

γ −
A,t
where all terms on the right hand side are constant, except for ln m

and γ.
6.1. ALTERNATIVE AND NON-STATIONARY EXAMPLE 71
initial condition in the case of a deterministic trend, since the variance is finite.
6.1.10 Specifying the steady state
Declaring the steady state is just as explained in details and according to
the same syntax explained in chapter 3, covering the initval, steady and
check commands. In chapter 5, section 5.5, we also discussed the usefulness
of providing an external Matlab file to solve for the steady state. In this
case, you can find the corresponding steady state file in the models folder
under UserGuide. The file is called fs2000ns steadystate.m. There are some
things to notice. First, the output of the function is the endogenous variables
at steady state, the ys vector. The check=0 limits steady state values to
real numbers. Second, notice the declaration of parameters at the beginning;
intuitive, but tedious... This functionality may be updated in later versions of
Dynare. Third, note that the file is really only a sequential set of equalities,
defining each variable in terms of parameters or variables solved in the lines
above. So far, nothing has changed with respect to the equivalent file of
chapter 5. The only novelty is the declaration of the non-stationary variables,
P obs and Y obs which take the value of 1. This is Dynare convention and
must be the case for all your non-stationary variables.
6.1.11 Declaring priors
We expand our .mod file with the following information:
estimated params;
alp, beta pdf, 0.356, 0.02;
bet, beta pdf, 0.993, 0.002;
gam, normal pdf, 0.0085, 0.003;
mst, normal pdf, 1.0002, 0.007;
rho, beta pdf, 0.129, 0.223;
psi, beta pdf, 0.65, 0.05;
del, beta pdf, 0.01, 0.005;
stderr e a, inv gamma pdf, 0.035449, inf;
stderr e m, inv gamma pdf, 0.008862, inf;
end;
6.1.12 Launching the estimation
We add the following commands to ask Dynare to run a basic estimation of
our model:
72 CHAPTER 6. ESTIMATING DSGE MODELS - ADVANCED TOPICS
estimation(datafile=fsdat,nobs=192,loglinear,mh replic=2000,
mode compute=6,mh nblocks=2,mh drop=0.45,mh jscale=0.65);
NOTE! As mentioned earlier, we need to instruct Dynare to log-linearize
our model, since it contains non-linear equations in non-stationary variables.
A simple linearization would fail as these variables do not have a steady state.
Fortunately, taking the log of the equations involving non-stationary variables
does the job of linearizing them.
6.1.13 The complete .mod file
We have seen each part of the .mod separately; it’s now time to get a picture
of what the complete file looks like. For convenience, the file also appears in
the models folder under UserGuide in your Dynare installation. The file is
called fs2000ns.mod.
var m P c e W R k d n l Y obs P obs y dA;
varexo e a e m;
parameters alp, bet, gam, mst, rho, psi, del;
model;
dA = exp(gam+e a);
log(m) = (1-rho)*log(mst) + rho*log(m(-1))+e m;
-P/(c(+1)*P(+1)*m)+bet*P(+1)*(alp*exp(-alp*(gam+log(e(+1))))
*k^(alp-1)*n(+1)^(1-alp)+(1-del)
*exp(-(gam+log(e(+1)))))/(c(+2)*P(+2)*m(+1))=0;
W = l/n;
-(psi/(1-psi))*(c*P/(1-n))+l/n = 0;
R = P*(1-alp)*exp(-alp*(gam+e a))*k(-1)^alp*n^(-alp)/W;
1/(c*P)-bet*P*(1-alp)*exp(-alp*(gam+e a))*k(-1)^alp*n^(1-alp)/
(m*l*c(+1)*P(+1)) = 0;
c+k = exp(-alp*(gam+e a))*k(-1)^alp*n^(1-alp)+(1-del)
exp(-(gam+e a))*k(-1);
P*c = m;
m-1+d = l;
e = exp(e a);
y = k(-1)^alp*n^(1-alp)*exp(-alp*(gam+e a));
Y obs/Y obs(-1) = dA*y/y(-1);
P obs/P obs(-1) = (p/p(-1))*m(-1)/dA;
end;
varobs P obs Y obs;
6.1. ALTERNATIVE AND NON-STATIONARY EXAMPLE 73
observation trends;
P obs (log(mst)-gam);
Y obs (gam);
end;
unit root vars P obs Y obs;
initval;
k = 6;
m = mst;
P = 2.25;
c = 0.45;
e = 1;
W = 4;
R = 1.02;
d = 0.85;
n = 0.19;
l = 0.86;
y = 0.6;
dA = exp(gam);
end;
// the above is really only useful if you want to do a stoch simul
// of your model, since the estimation will use the Matlab
// steady state file also provided and discussed above.
steady;
estimated params;
alp, beta pdf, 0.356, 0.02;
bet, beta pdf, 0.993, 0.002;
gam, normal pdf, 0.0085, 0.003;
mst, normal pdf, 1.0002, 0.007;
rho, beta pdf, 0.129, 0.223;
psi, beta pdf, 0.65, 0.05;
del, beta pdf, 0.01, 0.005;
stderr e a, inv gamma pdf, 0.035449, inf;
stderr e m, inv gamma pdf, 0.008862, inf;
end;
estimation(datafile=fsdat,nobs=192,loglinear,mh replic=2000,
mode compute=6,mh nblocks=2,mh drop=0.45,mh jscale=0.65);
74 CHAPTER 6. ESTIMATING DSGE MODELS - ADVANCED TOPICS
Figure 6.2: At a high level, there are five basic steps to translate a model into
Dynare for successful estimation.
6.1.14 Summing it up
The explanations given above of each step necessary to translate the Schorfheide
(2000) example into language that Dynare can understand and process was
quite lengthy and involved a slew of new commands and information. It may
therefore be useful, to gain a “bird’s eyeview” on what we have just accom-
plished, and summarize the most important steps at a high level. This is done
in figure 6.1.14.
6.2 Comparing models based on their posterior
distributions
** TBD
6.3. WHERE IS YOUR OUTPUT STORED? 75
6.3 Where is your output stored?
The output from estimation can be extremely varied, depending on the in-
structions you give Dynare. The Reference Manual overviews the complete
set of potential output files and describes where you can find each one.
Chapter 7
Solving DSGE models -
Behind the scenes of Dynare
7.1 Introduction
The aim of this chapter is to peer behind the scenes of Dynare, or under its
hood, to get an idea of the methodologies and algorithms used in its com-
putations. Going into details would be beyond the scope of this User Guide
which will instead remain at a high level. What you will find below will
either comfort you in realizing that Dynare does what you expected of it -
and what you would have also done if you had had to code it all yourself
(with a little extra time on your hands!), or will spur your curiosity to have
a look at more detailed material. If so, you may want to go through Michel
Juillard’s presentation on solving DSGE models to a first and second order
(available on Michel Juillard’s website), or read Collard and Juillard (2001a)
or Schmitt-Grohe and Uribe (2004) which gives a good overview of the most
recent solution techniques based on perturbation methods. Finally, note that
in this chapter we will focus on stochastic models - which is where the major
complication lies, as explained in section 3.1.1 of chapter 3. For more details
on the Newton-Raphson algorithm used in Dynare to solve deterministic mod-
els, see Juillard (1996).
7.2 What is the advantage of a second order
approximation?
As noted in chapter 3 and as will become clear in the section below, lin-
earizing a system of equations to the first order raises the issue of certainty
equivalence. This is because only the first moments of the shocks enter the
linearized equations, and when expectations are taken, they disappear. Thus,
77
78
CHAPTER 7. SOLVING DSGE MODELS - BEHIND THE SCENES OF
DYNARE
unconditional expectations of the endogenous variables are equal to their non-
stochastic steady state values.
This may be an acceptable simplification to make. But depending on the
context, it may instead be quite misleading. For instance, when using sec-
ond order welfare functions to compare policies, you also need second order
approximations of the policy function. Yet more clearly, in the case of asset
pricing models, linearizing to the second order enables you to take risk (or the
variance of shocks) into consideration - a highly desirable modeling feature. It
is therefore very convenient that Dynare allows you to choose between a first
or second order linearization of your model in the option of the stoch simul
command.
7.3 How does dynare solve stochastic DSGE
models?
In this section, we shall briefly overview the perturbation methods employed
by Dynare to solve DSGE models to a first order approximation. The sec-
ond order follows very much the same approach, although at a higher level
of complexity. The summary below is taken mainly from Michel Juillard’s
presentation “Computing first order approximations of DSGE models with
Dynare”, which you should read if interested in particular details, especially
regarding second order approximations (available on Michel Juillard’s web-
site).
To summarize, a DSGE model is a collection of first order and equilibrium
conditions that take the general form:
E
t
{f(y
t+1
, y
t
, y
t−1
, u
t
)} = 0
E(u
t
) = 0
E(u
t
u

t
) = Σ
u
and where:
y : vector of endogenous variables of any dimension
u : vector of exogenous stochastic shocks of any dimension
The solution to this system is a set of equations relating variables in the
current period to the past state of the system and current shocks, that satisfy
the original system. This is what we call the policy function. Sticking to the
above notation, we can write this function as:
y
t
= g(y
t−1
, u
t
)
7.3. HOW DOES DYNARE SOLVE STOCHASTIC DSGE MODELS? 79
Then, it is straightforward to re-write y
t+1
as
y
t+1
= g(y
t
, u
t+1
)
= g(g(y
t−1
, u
t
), u
t+1
)
We can then define a new function F, such that:
F(y
t−1
, u
t
, u
t+1
) = f(g(g(y
t−1
, u
t
), u
t+1
), g(y
t−1
, u
t
), y
t−1
, u
t
)
which enables us to rewrite our system in terms of past variables, and current
and future shocks:
E
t
[F(y
t−1
, u
t
, u
t+1
)] = 0
We then venture to linearize this model around a steady state defined as:
f(¯ y, ¯ y, ¯ y, 0) = 0
having the property that:
¯ y = g(¯ y, 0)
The first order Taylor expansion around ¯ y yields:
E
t
_
F
(1)
(y
t−1
, u
t
, u
t+1
)
_
=
E
t
_
f(¯ y, ¯ y, ¯ y) +f
y
+
_
g
y
(g
y
ˆ y +g
u
u) +g
u
u

_
+f
y
0
(g
y
ˆ y +g
u
u) +f
y

ˆ y +f
u
u
_
= 0
with ˆ y = y
t−1
− ¯ y, u = u
t
, u

= u
t+1
, f
y
+
=
∂f
∂y
t+1
, f
y
0
=
∂f
∂y
t
, f
y

=
∂f
∂y
t−1
,
f
u
=
∂f
∂u
t
, g
y
=
∂g
∂y
t−1
, g
u
=
∂g
∂u
t
.
Taking expectations (we’re almost there!):
E
t
_
F
(1)
(y
t−1
, u
t
, u
t+1
)
_
=
f(¯ y, ¯ y, ¯ y) +f
y
+
(g
y
(g
y
ˆ y +g
u
u))
+f
y
0
(g
y
ˆ y +g
u
u) +f
y

ˆ y +f
u
u
_
=
_
f
y
+
g
y
g
y
+f
y
0
g
y
+f
y

_
ˆ y +
_
f
y
+
g
y
g
u
+f
y
0
g
u
+f
u
_
u
= 0
As you can see, since future shocks only enter with their first moments
(which are zero in expectations), they drop out when taking expectations of
the linearized equations. This is technically why certainty equivalence holds
80
CHAPTER 7. SOLVING DSGE MODELS - BEHIND THE SCENES OF
DYNARE
in a system linearized to its first order. The second thing to note is that we
have two unknown variables in the above equation: g
y
and g
u
each of which
will help us recover the policy function g.
Since the above equation holds for any ˆ y and any u, each parenthesis must
be null and we can solve each at a time. The first, yields a quadratic equation
in g
y
, which we can solve with a series of algebraic trics that are not all imme-
diately apparent (but detailed in Michel Juillard’s presentation). Incidentally,
one of the conditions that comes out of the solution of this equation is the
Blanchard-Kahn condition: there must be as many roots larger than one in
modulus as there are forward-looking variables in the model. Having recov-
ered g
y
, recovering g
u
is then straightforward from the second parenthesis.
Finally, notice that a first order linearization of the function g yields:
y
t
= ¯ y +g
y
ˆ y +g
u
u
And now that we have g
y
and g
u
, we have solved for the (approximate) policy
(or decision) function and have succeeded in solving our DSGE model. If we
were interested in impulse response functions, for instance, we would simply
iterate the policy function starting from an initial value given by the steady
state.
The second order solution uses the same “perturbation methods” as above
(the notion of starting from a function you can solve - like a steady state -
and iterating forward), yet applies more complex algebraic techniques to re-
cover the various partial derivatives of the policy function. But the general
approach is perfectly isomorphic. Note that in the case of a second order
approximation of a DSGE model, the variance of future shocks remains after
taking expectations of the linearized equations and therefore affects the level
of the resulting policy function.
Chapter 8
Estimating DSGE models -
Behind the scenes of Dynare
This chapter focuses on the theory of Bayesian estimation. It begins by mo-
tivating Bayesian estimation by suggesting some arguments in favor of it as
opposed to other forms of model estimation. It then attempts to shed some
light on what goes on in Dynare’s machinery when it estimates DSGE models.
To do so, this section surveys the methodologies adopted for Bayesian estima-
tion, including defining what are prior and posterior distributions, using the
Kalman filter to find the likelihood function, estimating the posterior function
thanks to the Metropolis-Hastings algorithm, and comparing models based on
posterior distributions.
8.1 Advantages of Bayesian estimation
Bayesian estimation is becoming increasingly popular in the field of macro-
economics. Recent papers have attracted significant attention; some of these
include: Schorfheide (2000) which uses Bayesian methods to compare the
fit of two competing DSGE models of consumption, Lubik and Schorfheide
(2003) which investigates whether central banks in small open economies re-
spond to exchange rate movements, Smets and Wouters (2003) which ap-
plies Bayesian estimation techniques to a model of the Eurozone, Ireland
(2004) which emphasizes instead maximum likelihood estimation, Fernandez-
Villaverde and Rubio-Ramirez (2004) which reviews the econometric proper-
ties of Bayesian estimators and compare estimation results with maximum
likelihood and BVAR methodologies, Lubik and Schorfheide (2005) which ap-
plies Bayesian estimation methods to an open macro model focussing on issues
of misspecification and identification, and finally Rabanal and Rubio-Ramirez
(2005) which compares the fit, based on posterior distributions, of four com-
peting specifications of New Keynesian monetary models with nominal rigidi-
ties.
81
82
CHAPTER 8. ESTIMATING DSGE MODELS - BEHIND THE SCENES
OF DYNARE
There are a multitude of advantages of using Bayesian methods to esti-
mate a model, but five of these stand out as particularly important and general
enough to mention here.
First, Bayesian estimation fits the complete, solved DSGE model, as op-
posed to GMM estimation which is based on particular equilibrium relation-
ships such as the Euler equation in consumption. Likewise, estimation in the
Bayesian case is based on the likelihood generated by the DSGE system, rather
than the more indirect discrepancy between the implied DSGE and VAR im-
pulse response functions. Of course, if your model is entirely mis-specified,
estimating it using Bayesian techniques could be a disadvantage.
Second, Bayesian techniques allow the consideration of priors which work
as weights in the estimation process so that the posterior distribution avoids
peaking at strange points where the likelihood peaks. Indeed, due to the
stylized and often misspecified nature of DSGE models, the likelihood often
peaks in regions of the parameter space that are contradictory with common
observations, leading to the “dilemma of absurd parameter estimates”.
Third, the inclusion of priors also helps identifying parameters. Unfortu-
nately, when estimating a model, the problem of identification often arises. It
can be summarized by different values of structural parameters leading to the
same joint distribution for observables. More technically, the problem arises
when the posterior distribution is flat over a subspace of parameter values.
But the weighting of the likelihood with prior densities often leads to adding
just enough curvature in the posterior distribution to facilitate numerical max-
imization.
Fourth, Bayesian estimation explicitly addresses model misspecification by
including shocks, which can be interpreted as observation errors, in the struc-
tural equations.
Sixth, Bayesian estimation naturally leads to the comparison of models
based on fit. Indeed, the posterior distribution corresponding to competing
models can easily be used to determine which model best fits the data. This
procedure, as other topics mentioned above, is discussed more technically in
the subsection below.
8.2 The basic mechanics of Bayesian estimation
This and the following subsections are based in great part on work by, and
discussions with, St´ephane Adjemian, a member of the Dynare development
8.2. THE BASIC MECHANICS OF BAYESIAN ESTIMATION 83
team. Some of this work, although summarized in presentation format, is
available in the “Events” page of the Dynare website. Other helpful material
includes An and Schorfheide (2006), which includes a clear and quite complete
introduction to Bayesian estimation, illustrated by the application of a sim-
ple DSGE model. Also, the appendix of Schorfheide (2000) contains details
as to the exact methodology and possible difficulties encountered in Bayesian
estimation. You may also want to take a glance at Hamilton (1994), chapter
12, which provides a very clear, although somewhat outdated, introduction
to the basic mechanics of Bayesian estimation. Finally, the websites of Frank
Schorfheide and Jesus Fernandez-Villaverde contain a wide variety of very
helpful material, from example files to lecture notes to related papers. Fi-
nally, remember to also check the open online examples of the Dynare website
for examples of .mod files touching on Bayesian estimation.
At its most basic level, Bayesian estimation is a bridge between calibra-
tion and maximum likelihood. The tradition of calibrating models is inherited
through the specification of priors. And the maximum likelihood approach en-
ters through the estimation process based on confronting the model with data.
Together, priors can be seen as weights on the likelihood function in order to
give more importance to certain areas of the parameter subspace. More tech-
nically, these two building blocks - priors and likelihood functions - are tied
together by Bayes’ rule. Let’s see how.
First, priors are described by a density function of the form
p(θ
A
|A)
where A stands for a specific model, θ
A
represents the parameters of model A,
p(•) stands for a probability density function (pdf) such as a normal, gamma,
shifted gamma, inverse gamma, beta, generalized beta, or uniform function.
Second, the likelihood function describes the density of the observed data,
given the model and its parameters:
L(θ
A
|Y
T
, A) ≡ p(Y
T

A
, A)
where Y
T
are the observations until period T, and where in our case the
likelihood is recursive and can be written as:
p(Y
T

A
, A) = p(y
0

A
, A)
T

t=1
p(y
t
|Y
t−1
, θ
A
, A)
We now take a step back. Generally speaking, we have a prior density p(θ)
on the one hand, and on the other, a likelihood p(Y
T
|θ). In the end, we are
interested in p(θ|Y
T
), the posterior density. Using the Bayes theorem
84
CHAPTER 8. ESTIMATING DSGE MODELS - BEHIND THE SCENES
OF DYNARE
twice we obtain this density of parameters knowing the data. Generally, we
have
p(θ|Y
T
) =
p(θ; Y
T
)
p(Y
T
)
We also know that
p(Y
T
|θ) =
p(θ; Y
T
)
p(θ)
⇔p(θ; Y
T
) = p(Y
T
|θ) ×p(θ)
By using these identities, we can combine the prior density and the
likelihood function discussed above to get the posterior density:
p(θ
A
|Y
T
, A) =
p(Y
T

A
, A)p(θ
A
|A)
p(Y
T
|A)
where p(Y
T
|A) is the marginal density of the data conditional on the model:
p(Y
T
|A) =
_
Θ
A
p(θ
A
; Y
T
|A)dθ
A
Finally, the posterior kernel (or un-normalized posterior density, given
that the marginal density above is a constant or equal for any parameter),
corresponds to the numerator of the posterior density:
p(θ
A
|Y
T
, A) ∝ p(Y
T

A
, A)p(θ
A
|A) ≡ K(θ
A
|Y
T
, A)
This is the fundamental equation that will allow us to rebuild all posterior mo-
ments of interest. The trick will be to estimate the likelihood function with
the help of the Kalman filter and then simulate the posterior kernel using
a sampling-like or Monte Carlo method such as the Metropolis-Hastings.
These topics are covered in more details below. Before moving on, though,
the subsection below gives a simple example based on the above reasoning
of what we mean when we say that Bayesian estimation is “somewhere in
between calibration and maximum likelihood estimation”. The example is
drawn from Zellner (1971), although other similar examples can be found in
Hamilton (1994), chapter 12.
8.2.1 Bayesian estimation: somewhere between calibration
and maximum likelihood estimation - an example
Suppose a data generating process y
t
= µ + ε
t
for t = 1, ..., T, where ε
t

N(0, 1) is gaussian white noise. Then, the likelihood is given by
p(Y
T
|µ) = (2π)

T
2
e

1
2

T
t=1
(y
t
−µ)
2
We know from the above that ´ µ
ML,T
=
1
T

T
t=1
y
t
≡ y and that V[´ µ
ML,T
] =
1
T
.
8.3. DSGE MODELS AND BAYESIAN ESTIMATION 85
In addition, let our prior be a gaussian distribution with expectation µ
0
and variance σ
2
µ
. Then, the posterior density is defined, up to a constant, by:
p (µ|Y
T
) ∝ (2πσ
2
µ
)

1
2
e

1
2
(µ−µ
0
)
2
σ
2
µ
×(2π)

T
2
e

1
2

T
t=1
(y
t
−µ)
2
Or equivalently, p (µ|Y
T
) ∝ e

(µ−E[µ])
2
V[µ]
, with
V[µ] =
1
_
1
T
_
−1

−2
µ
and
E[µ] =
_
1
T
_
−1
´ µ
ML,T

−2
µ
µ
0
_
1
T
_
−1

−2
µ
From this, we can tell that the posterior mean is a convex combination of
the prior mean and the ML estimate. In particular, if σ
2
µ
→ ∞ (ie, we have
no prior information, so we just estimate the model) then E[µ] → ´ µ
ML,T
, the
maximum likelihood estimator. But if σ
2
µ
→0 (ie, we’re sure of ourselves and
we calibrate the parameter of interest, thus leaving no room for estimation)
then E[µ] → µ
0
, the prior mean. Most of the time, we’re somewhere in the
middle of these two extremes.
8.3 DSGE models and Bayesian estimation
8.3.1 Rewriting the solution to the DSGE model
Recall from chapter 7 that any DSGE model, which is really a collection of first
order and equilibrium conditions, can be written in the formE
t
{f(y
t+1
, y
t
, y
t−1
, u
t
)} =
0, taking as a solution equations of the type y
t
= g(y
t−1
, u
t
), which we call
the decision rule. In more appropriate terms for what follows, we can rewrite
the solution to a DSGE model as a system in the following manner:
y

t
= M¯ y(θ) +Mˆ y
t
+N(θ)x
t

t
ˆ y
t
= g
y
(θ)ˆ y
t−1
+g
u
(θ)u
t
E(η
t
η

t
) = V (θ)
E(u
t
u

t
) = Q(θ)
where ˆ y
t
are variables in deviations from steady state, ¯ y is the vector of steady
state values and θ the vector of deep (or structural) parameters to be esti-
mated. Other variables are described below.
The second equation is the familiar decision rule mentioned above. But
the equation expresses a relationship among true endogenous variables that
are not directly observed. Only y

t
is observable, and it is related to the true
86
CHAPTER 8. ESTIMATING DSGE MODELS - BEHIND THE SCENES
OF DYNARE
variables with an error η
t
. Furthermore, it may have a trend, which is captured
with N(θ)x
t
to allow for the most general case in which the trend depends on
the deep parameters. The first and second equations above therefore naturally
make up a system of measurement and transition or state equations, respec-
tively, as is typical for a Kalman filter (you guessed it, it’s not a coincidence!).
8.3.2 Estimating the likelihood function of the DSGE model
The next logical step is to estimate the likelihood of the DSGE solution system
mentioned above. The first apparent problem, though, is that the equations
are non linear in the deep parameters. Yet, they are linear in the endogenous
and exogenous variables so that the likelihood may be evaluated with a linear
prediction error algorithm like the Kalman filter. This is exactly what Dynare
does. As a reminder, here’s what the Kalman filter recursion does.
For t = 1, . . . , T and with initial values y
1
and P
1
given, the recursion
follows
v
t
= y

t
− ¯ y

−Mˆ y
t
−Nx
t
F
t
= MP
t
M

+V
K
t
= g
y
P
t
g

y
F
−1
t
ˆ y
t+1
= g
y
ˆ y
t
+K
t
v
t
P
t+1
= g
y
P
t
(g
y
−K
t
M)

+g
u
Qg

u
For more details on the Kalman filter, see Hamilton (1994), chapter 13.
From the Kalman filter recursion, it is possible to derive the log-likelihood
given by
ln L(θ|Y

T
) = −
Tk
2
ln(2π) −
1
2
T

t=1
|F
t
| −
1
2
v

t
F
−1
t
v
t
where the vector θ contains the parameters we have to estimate: θ, V (θ) and
Q(θ) and where Y

T
expresses the set of observable endogenous variables y

t
found in the measurement equation.
The log-likelihood above gets us one step closer to our goal of finding the
posterior distribution of our parameters. Indeed, the log posterior kernel
can be expressed as
ln K(θ|Y

T
) = ln L(θ|Y

T
) + ln p(θ)
where the first term on the right hand side is now known after carrying out the
Kalman filter recursion. The second, recall, are the priors, and are also known.
8.3. DSGE MODELS AND BAYESIAN ESTIMATION 87
8.3.3 Finding the mode of the posterior distribution
Next, to find the mode of the posterior distribution - a key parameter and
an important output of Dynare - we simply maximize the above log posterior
kernel with respect to θ. This is done in Dynare using numerical methods.
Recall that the likelihood function is not Gaussian with respect to θ but to
functions of θ as they appear in the state equation. Thus, this maximization
problem is not completely straightforward, but fortunately doable with mod-
ern computers.
8.3.4 Estimating the posterior distribution
Finally, we are now in a position to find the posterior distribution of our
parameters. The distribution will be given by the kernel equation above,
but again, it is a nonlinear and complicated function of the deep parameters
θ. Thus, we cannot obtain an explicit form for it. We resort, instead, to
sampling-like methods, of which the Metropolis-Hastings has been retained in
the literature as particularly efficient. This is indeed the method adopted by
Dynare.
The general idea of the Metropolis-Hastings algorithm is to simulate the
posterior distribution. It is a “rejection sampling algorithm” used to generate
a sequence of samples (also known as a “Markov Chain” for reasons that will
become apparent later) from a distribution that is unknown at the outset.
Remember that all we have is the posterior mode; we are instead more often
interested in the mean and variance of the estimators of θ. To do so, the
algorithm builds on the fact that under general conditions the distribution
of the deep parameters will be asymptotically normal. The algorithm, in the
words of An and Shorfheide, “constructs a Gaussian approximation around
the posterior mode and uses a scaled version of the asymptotic covariance
matrix as the covariance matrix for the proposal distribution. This allows for
an efficient exploration of the posterior distribution at least in the neighbor-
hood of the mode” (An and Schorfheide (2006), p. 18). More precisely, the
Metropolis-Hastings algorithm implements the following steps:
1. Choose a starting point θ

, where this is typically the posterior mode,
and run a loop over 2-3-4.
2. Draw a proposal θ

from a jumping distribution
J(θ


t−1
) = N(θ
t−1
, cΣ
m
)
where Σ
m
is the inverse of the Hessian computed at the posterior mode.
88
CHAPTER 8. ESTIMATING DSGE MODELS - BEHIND THE SCENES
OF DYNARE
3. Compute the acceptance ratio
r =
p(θ

|Y
T
)
p(θ
t−1
|Y
T
)
=
K(θ

|Y
T
)
K(θ
t−1
|Y
T
)
4. Finally accept or discard the proposal θ

according to the following rule,
and update, if necessary, the jumping distribution:
θ
t
=
_
θ

with probability min(r, 1)
θ
t−1
otherwise.
Figure 8.3.4 tries to clarify the above. In step 1, choose a candidate
paramter, θ

from a Normal distribution, whose mean has been set to θ
t−1
(this will become clear in just a moment). In step 2, compute the value of
the posterior kernel for that candidate parameter, and compare it to the value
of the kernel from the mean of the drawing distribution. In step 3, decide
whether or not to hold on to your candidate parameter. If the acceptance
ratio is greater than one, then definitely keep your candidate. Otherwise, go
back to the candidate of last period (this is true in very coarse terms, notice
that in fact you would keep your candidate only with a probability less than
one). Then, do two things. Update the mean of your drawing distribution,
and note the value of the parameter your retain. After having repeated these
steps often enough, in the final step, build a histogram of those retained val-
ues. Of course, the point is for each “bucket” of the histogram to shrink to
zero. This “smoothed histogram” will eventually be the posterior distribution
after sufficient iterations of the above steps.
But why have such a complicated acceptance rule? The point is to be able
to visit the entire domain of the posterior distribution. We should not be too
quick to simply throw out the candidate giving a lower value of the posterior
kernel, just in case using that candidate for the mean of the drawing distri-
bution allows us to to leave a local maximum and travel towards the global
maximum. Metaphorically, the idea is to allow the search to turn away from
taking a small step up, and instead take a few small steps down in the hope
of being able to take a big step up in the near future. Of course, an important
parameter in this searching procedure is the variance of the jumping distri-
bution and in particular the scale factor. If the scale factor is too small,
the acceptance rate (the fraction of candidate parameters that are accepted
in a window of time) will be too high and the Markov Chain of candidate
parameters will “mix slowly”, meaning that the distribution will take a long
time to converge to the posterior distribution since the chain is likely to get
“stuck” around a local maximum. On the other hand, if the scale factor is
too large, the acceptance rate will be very low (as the candidates are likely to
land in regions of low probability density) and the chain will spend too much
8.3. DSGE MODELS AND BAYESIAN ESTIMATION 89
Figure 8.1: The above sketches the Metropolis-Hastings algorithm, used to
build the posterior distribution function. Imagine repeating these steps a large
number of times, and smoothing the “histogram” such that each “bucket” has
infinitely small width.
time in the tails of the posterior distribution.
While these steps are mathematically clear, at least to a machine needing
to undertake the above calculations, several practical questions arise when
carrying out Bayesian estimation. These include: How should we choose the
scale factor c (variance of the jumping distribution)? What is a satisfactory
acceptance rate? How many draws are ideal? How is convergence of the
Metropolis-Hastings iterations assessed? These are all important questions
that will come up in your usage of Dynare. They are addressed as clearly as
possible in section 5.7 of Chapter 5.
90
CHAPTER 8. ESTIMATING DSGE MODELS - BEHIND THE SCENES
OF DYNARE
8.4 Comparing models using posterior
distributions
As mentioned earlier, while touting the advantages of Bayesian estimation,
the posterior distribution offers a particularly natural method of comparing
models. Let’s look at an illustration.
Suppose we have a prior distribution over two competing models: p(A)
and p(B). Using Bayes’ rule, we can compute the posterior distribution over
models, where I = A, B
p(I|Y
T
) =
p(I)p(Y
T
|I)

I=A,B
p(I)p(Y
T
|I)
where this formula may easily be generalized to a collection of N models.
Then, the comparison of the two models is done very naturally through the
ratio of the posterior model distributions. We call this the posterior odds
ratio:
p(A|Y
T
)
p(B|Y
T
)
=
p(A)
p(B)
p(Y
T
|A)
p(Y
T
|B)
The only complication is finding the magrinal density of the data condi-
tional on the model, p(Y
T
|I), which is also the denominator of the posterior
density p(θ|Y
T
) discussed earlier. This requires some detailed explanations
of their own.
For each model I = A, B we can evaluate, at least theoretically, the
marginal density of the data conditional on the model by integrating out
the deep parameters θ
I
from the posterior kernel:
p(Y
T
|I) =
_
Θ
I
p(θ
I
; Y
T

I
, I)dθ
I
=
_
Θ
I
p(θ
I
|I) ×p(Y
T

I
, I)dθ
I
Note that the expression inside the integral sign is exactly the posterior kernel.
To remind you of this, you may want to glance back at the first subsection
above, specifying the basic mechanics of Bayesian estimation.
To obtain the marginal density of the data conditional on the model, there
are two options. The first is to assume a functional form of the posterior kernel
that we can integrate. The most straightforward and the best approximation,
especially for large samples, is the Gaussian (called a Laplace approxima-
tion). In this case, we would have the following estimator:
´ p(Y
T
|I) = (2π)
k
2

θ
m
I
|
1
2
p(θ
m
I
|Y
T
, I)p(θ
m
I
|I)
where θ
m
I
is the posterior mode. The advantage of this technique is its com-
putational efficiency: time consuming Metropolis-Hastings iterations are not
8.4. COMPARING MODELS USING POSTERIOR DISTRIBUTIONS 91
necessary, only the numerically calculated posterior mode is required.
The second option is instead to use information from the Metropolis-
Hastings runs and is typically referred to as the Harmonic Mean Esti-
mator. The idea is to simulate the marginal density of interest and to simply
take an average of these simulated values. To start, note that
p(Y
T
|I) = E
_
f(θ
I
)
p(θ
I
|I)p(Y
T

I
, I)
¸
¸
¸
¸
θ
I
, I
_
−1
where f is a probability density function, since
E
_
f(θ
I
)
p(θ
I
|I)p(Y
T

I
, I)
¸
¸
¸
¸
θ
I
, I
_
=
_
Θ
I
f(θ)dθ
_
Θ
I
p(θ
I
|I)p(Y
T

I
, I)dθ
I
and the numerator integrates out to one (seeGeweke (1999) for more details).
This suggests the following estimator of the marginal density
´ p(Y
T
|I) =
_
1
B
B

b=1
f(θ
(b)
I
)
p(θ
(b)
I
|I)p(Y
T

(b)
I
, I)
_
−1
where each drawn vector θ
(b)
I
comes from the Metropolis-Hastings iterations
and where the probability density function f can be viewed as a weights on
the posterior kernel in order to downplay the importance of extreme values of
θ. Geweke (1999) suggests to use a truncated Gaussian function, leading to
what is typically referred to as the Modified Harmonic Mean Estimator.
Chapter 9
Optimal policy under
commitment
93
Chapter 10
Troubleshooting
To make sure this section is as user friendly as possible, the best is to compile
what users have to say! Please let me know what your most common problem
is with Dynare, how Dynare tells you about it and how you solve it. Thanks
for your precious help!
95
Bibliography
An, S., and F. Schorfheide (2006): “Bayesian Analysis of DSGE Models,”
Econometric Review, Forthcoming.
Clarida, R., J. Gali, and M. Gertler (1999): “The Science of Monetary
Policy: A New Keynesian Perspective,” Journal of Economic Literature,
XXXVII, 1661–1707.
Collard, F., and M. Juillard (2001a): “Accuracy of stochastic pertur-
bation methods: The case of asset pricing models,” Journal of Economic
Dynamics and Control, 25(6-7), 979–999.
(2001b): “A Higher-Order Taylor Expansion Approach to Simulation
of Stochastic Forward-Looking Models with an Application to a Nonlinear
Phillips Curve Model,” Computational Economics, 17(2-3), 125–39.
(2003): “Stochastic simulations with DYNARE. A practical guide.,”
CEPREMAP mimeo.
Durbin, J., and S. Koopman (2001): Time Series Analysis by State Space
Methods. Oxford University Press, Oxford, U.K.
Fernandez-Villaverde, J., and J. F. Rubio-Ramirez (2004): “Compar-
ing dynamic equilibrium models to data: a Bayesian approach,” Journal of
Econometrics, 123(1), 153–187.
Geweke, J. (1999): “Using Simulation Methods for Bayesian Econometric
Models: In- ference, Development and Communication,” Econometric Re-
view, 18(1), 1–126.
Hamilton, J. D. (1994): Time Series Analysis. Princeton University Press,
Princeton, NJ.
Ireland, P. N. (2004): “A method for taking models to the data,” Journal
of Economic Dynamics and Control, 28(6), 1205–1226.
Juillard, M. (1996): “Dynare : a program for the resolution and simulation
of dynamic models with forward variables through the use of a relaxation
algorithm,” CEPREMAP working papers 9602, CEPREMAP.
97
98 BIBLIOGRAPHY
Lubik, T., and F. Schorfheide (2003): “Do Central Banks Respond to
Exchange Rate Movements? A Structural Investigation,” Economics Work-
ing Paper Archive 505, The Johns Hopkins University,Department of Eco-
nomics.
(2005): “A Bayesian Look at New Open Economy Macroeco-
nomics,” Economics Working Paper Archive 521, The Johns Hopkins Uni-
versity,Department of Economics.
Nason, J. M., and T. Cogley (1994): “Testing the Implications of Long-
Run Neutrality for Monetary Business Cycle Models,” Journal of Applied
Econometrics, 9(S), S37–70.
Rabanal, P., and J. F. Rubio-Ramirez (2005): “Comparing New Key-
nesian models of the business cycle: A Bayesian approach,” Journal of
Monetary Economics, 52(6), 1151–1166.
Schmitt-Grohe, S., and M. Uribe (2004): “Solving dynamic general equi-
librium models using a second-order approximation to the policy function,”
Journal of Economic Dynamics and Control, 28(4), 755–775.
Schorfheide, F. (2000): “Loss function-based evaluation of DSGE models,”
Journal of Applied Econometrics, 15(6), 645–670.
Smets, F., and R. Wouters (2003): “An Estimated Dynamic Stochastic
General Equilibrium Model of the Euro Area,” Journal of the European
Economic Association, 1(5), 1123–1175.
Zellner, A. (1971): An Introduction to Bayesian Inference in Econometrics.
John Wiley & Sons, Inc., New York.

Dynare v4 - User Guide Public beta version

Tommaso Mancini Griffoli tommaso.mancini@stanfordalumni.org This draft: June 2010

iii

Copyright c 2007-2010 Tommaso Mancini Griffoli Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license can be found at: http://www.gnu.org/licenses/ fdl.txt

Contents
Contents List of Figures 1 Introduction 1.1 About this Guide - approach and structure 1.2 What is Dynare? . . . . . . . . . . . . . . . 1.3 Additional sources of help . . . . . . . . . . 1.4 Nomenclature . . . . . . . . . . . . . . . . . 1.5 v4, what’s new and backward compatibility 2 Installing Dynare 2.1 Dynare versions . . . . . 2.2 System requirements . . 2.3 Installing Dynare . . . . 2.4 MATLAB particularities iv vii 1 1 2 4 5 5 7 7 7 7 8 9 9 10 11 15 15 16 16 17 18 18 19 19 19 20 20 21

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

3 Solving DSGE models - basics 3.1 A fundamental distinction . . . . . . . . . . . . . . . . . . . 3.1.1 NOTE! Deterministic vs stochastic models . . . . . . 3.2 Introducing an example . . . . . . . . . . . . . . . . . . . . 3.3 Dynare .mod file structure . . . . . . . . . . . . . . . . . . . 3.4 Filling out the preamble . . . . . . . . . . . . . . . . . . . . 3.4.1 The deterministic case . . . . . . . . . . . . . . . . . 3.4.2 The stochastic case . . . . . . . . . . . . . . . . . . . 3.4.3 Comments on your first lines of Dynare code . . . . 3.5 Specifying the model . . . . . . . . . . . . . . . . . . . . . . 3.5.1 Model in Dynare notation . . . . . . . . . . . . . . . 3.5.2 General conventions . . . . . . . . . . . . . . . . . . 3.5.3 Notational conventions . . . . . . . . . . . . . . . . . 3.5.4 Timing conventions . . . . . . . . . . . . . . . . . . 3.5.5 Conventions specifying non-predetermined variables 3.5.6 Linear and log-linearized models . . . . . . . . . . . 3.6 Specifying steady states and/or initial values . . . . . . . . iv

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

CONTENTS 3.6.1 Stochastic models and steady states . . . . 3.6.2 Deterministic models and initial values . . . 3.6.3 Finding a steady state . . . . . . . . . . . . 3.6.4 Checking system stability . . . . . . . . . . 3.7 Adding shocks . . . . . . . . . . . . . . . . . . . . 3.7.1 Deterministic models - temporary shocks . 3.7.2 Deterministic models - permanent shocks . 3.7.3 Stochastic models . . . . . . . . . . . . . . 3.8 Selecting a computation . . . . . . . . . . . . . . . 3.8.1 For deterministic models . . . . . . . . . . . 3.8.2 For stochastic models . . . . . . . . . . . . 3.9 The complete .mod file . . . . . . . . . . . . . . . . 3.9.1 The stochastic model . . . . . . . . . . . . . 3.9.2 The deterministic model (case of temporary 3.10 File execution and results . . . . . . . . . . . . . . 3.10.1 Results - stochastic models . . . . . . . . . 3.10.2 Results - deterministic models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . shock) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v 21 23 23 24 25 25 25 27 27 28 28 31 31 32 33 33 34 37 37 37 38 41 42 43 43 44 44 44 44 46 47 47 48 48 49 49 49 52 55 57 57 57

4 Solving DSGE models - advanced topics 4.1 Dynare features and functionality . . . . . . . . . . . . . . 4.1.1 Other examples . . . . . . . . . . . . . . . . . . . . 4.1.2 Alternative, complete example . . . . . . . . . . . 4.1.3 Finding, saving and viewing your output . . . . . . 4.1.4 Referring to external files . . . . . . . . . . . . . . 4.1.5 Infinite eigenvalues . . . . . . . . . . . . . . . . . . 4.2 Files created by Dynare . . . . . . . . . . . . . . . . . . . 4.3 Modeling tips . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1 Stationarizing your model . . . . . . . . . . . . . . 4.3.2 Expectations taken in the past . . . . . . . . . . . 4.3.3 Infinite sums . . . . . . . . . . . . . . . . . . . . . 4.3.4 Infinite sums with changing timing of expectations 5 Estimating DSGE models - basics 5.1 Introducing an example . . . . . . 5.2 Declaring variables and parameters 5.3 Declaring the model . . . . . . . . 5.4 Declaring observable variables . . . 5.5 Specifying the steady state . . . . 5.6 Declaring priors . . . . . . . . . . . 5.7 Launching the estimation . . . . . 5.8 The complete .mod file . . . . . . . 5.9 Interpreting output . . . . . . . . . 5.9.1 Tabular results . . . . . . . 5.9.2 Graphical results . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

6 Estimating DSGE models - advanced topics 6.1 Alternative and non-stationary example . . . . . . . . . 6.1.1 Introducing the example . . . . . . . . . . . . . . 6.1.2 Declaring variables and parameters . . . . . . . . 6.1.3 The origin of non-stationarity . . . . . . . . . . . 6.1.4 Stationarizing variables . . . . . . . . . . . . . . 6.1.5 Linking stationary variables to the data . . . . . 6.1.6 The resulting model block of the .mod file . . . . 6.1.7 Declaring observable variables . . . . . . . . . . . 6.1.8 Declaring trends in observable variables . . . . . 6.1.9 Declaring unit roots in observable variables . . . 6.1.10 Specifying the steady state . . . . . . . . . . . . 6.1.11 Declaring priors . . . . . . . . . . . . . . . . . . 6.1.12 Launching the estimation . . . . . . . . . . . . . 6.1.13 The complete .mod file . . . . . . . . . . . . . . . 6.1.14 Summing it up . . . . . . . . . . . . . . . . . . . 6.2 Comparing models based on their posterior distributions 6.3 Where is your output stored? . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

61 61 61 66 66 67 68 68 69 69 70 71 71 71 72 74 74 75

7 Solving DSGE models - Behind the scenes of Dynare 77 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 7.2 What is the advantage of a second order approximation? . . . . 77 7.3 How does dynare solve stochastic DSGE models? . . . . . . . . 78 8 Estimating DSGE models - Behind the scenes of Dynare 81 8.1 Advantages of Bayesian estimation . . . . . . . . . . . . . . . . 81 8.2 The basic mechanics of Bayesian estimation . . . . . . . . . . . 82 8.2.1 Bayesian estimation: somewhere between calibration and maximum likelihood estimation - an example . . . . . . 84 8.3 DSGE models and Bayesian estimation . . . . . . . . . . . . . . 85 8.3.1 Rewriting the solution to the DSGE model . . . . . . . 85 8.3.2 Estimating the likelihood function of the DSGE model . 86 8.3.3 Finding the mode of the posterior distribution . . . . . 87 8.3.4 Estimating the posterior distribution . . . . . . . . . . . 87 8.4 Comparing models using posterior distributions . . . . . . . . . 90 9 Optimal policy under commitment 10 Troubleshooting vi 93 95

Illustration of the Metropolis-Hastings algorithm . . . . CIA model illustration . . .List of Figures Bibliography vii 97 List of Figures 1. . . Structure of the . . . . . . a bird’s eyeview . . . . . . . . . . . . . . . . . . Steps of model estimation . . . . . . . . . . . . . . . . .1 6. . . . . . . . . . . .mod file . . . . . . .2 8.1 3. . . . . . .1 Dynare. . . . . . 3 16 62 74 89 .1 6. . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

Thanks very much for your patience and good ideas. First. or at times notes to you . are these clear? On the contrary. notes to myself or others of the Dynare development team.mancini@stanfordalumni.our readers . These are mostly placeholders for future work. Any such notes are marked with two stars (**). The second thing that a work in progress manuscript comes with is a few internal notes. are there certain parts that just click particularly well? How can others be improved? I’m very interested to get your feedback.Work in Progress! This is the second version of the Dynare User Guide which is still work in progress! This means two things.org. or preferably on the Dynare Documentation Forum available in the Dynare Forums. Please write either direclty to myself: tommaso. are there enough examples. ix .to highlight a feature not yet fully stable. please read this with a critical eye and send me comments! Are some areas unclear? Is anything plain wrong? Are some sections too wordy.

.

Kamenik.co.uk) • Marco Ratto (marco. Wouters. report bugs and suggest new features. Klein.ec. Currently. remember. the development of Dynare could not have come such a long ways withough an active community of users who continually pose questions.fr) • Ferhat Mihoubi (ferhat.mihoubi“AT”univ-evry. Sakata. F.fr) e Several parts of Dynare use or have strongly benefited from publicly available programs by G. The help of this community is gratefully acknowledged. Anderson. The email addresses above are provided in case you wish to contact any one of the authors of Dynare directly. Schorfheide.fr) • Michel Juillard (michel. P.adjemian“AT”ens. the development team of Dynare is composed of • St´phane Adjemian (stephane. C.ratto“AT”jrc. Soederlind and R.europa.bastani“AT”ens. Collard. P. S. Sims. F. Finally. L.juillard“AT”mjui.orangehome.Contacts and Credits Dynare was originally developed by Michel Juillard in Paris.fr) e • Houtan Bastani (houtan. Ingber. France.villemot“AT”ens.eu) • S´bastien Villemot (sebastien. O.fr) • George Perendia (george“AT”perendia. We nonetheless encourage you to first use the Dynare forums to ask your questions so that other users can benefit from them as well. almost no question is specific enough to interest just one person. and yours is not the exception! xi .

.

though. The sophisticated computer programmer. respectively.needing a powerful and flexible program to support and facilitate his or her research activities in a variety of fields. we recommend starting with chapters 3 and 5. To root this understanding more deeply. from getting started to implementing advanced features. graduate student or central banker .1 About this Guide . may not find this Guide sufficiently detailed. or the specialist of computational economics. To do so. To do 1 .like a professor. this Guide also gives some background on Dynare’s algorithms. methodologies and underlying theory. We recognize that the “advanced economist” may be either a beginning or intermediate user of Dynare. on the one hand.approach and structure This User Guide aims to help you master Dynare’s main functionalities. this Guide is structured around examples and offers practical advice. Thus. This Guide is written to accommodate both. a secondary function of this Guide is to serve as a basic primer on DSGE model solving and Bayesian estimation. thus emphasizing depth over breadth. This Guide will focus on the most common or useful features of the program. The idea is to get you to use 90% of the program well and then tell you where else to look if you’re interested in fine tuning or advanced customization. on the other. This Guide is written mainly for an advanced economist . which introduce the program’s basic features to solve (including running impulse response functions) and estimate DSGE models.Chapter 1 Introduction Welcome to Dynare! 1. If you’re new to Dynare.

• TIP! introduces advice to help you work more efficiently with Dynare or solve common problems. . these chapters lead you through a complete hands-on example. you will most likely find the advanced chapters. . in order to “learn by doing”. • NOTE! is used to draw your attention to particularly important information you should keep in mind when using Dynare. you will probably find yourself coming back to the User Guide to skim over some of the content in the advanced chapters to iron out details and potential complications you may run into. we’ve introduced two different markers throughout the Guide to help streamline your reading. besides breaking up content into short chapters. These chapters cover more advanced features of Dynare and more complicated usage scenarios. which we recommend following from A to Z. We also recognize that you probably have had repeated if not active exposure to programming and are likely to have a strong economic background. The presumption is that you would skip around these chapters to focus on the topics most applicable to your needs and curiosity. though. to solve. 4 and 6. more appropriate. Examples are therefore more concise and specific to each feature. .2 CHAPTER 1. you will know the crux of Dynare’s functionality and (hopefully!) feel comfortable using Dynare for your own work. Once you have read these two chapters. Thus. INTRODUCTION so.2 What is Dynare? Before we dive into the thick of the “trees”. just what is Dynare? Dynare is a powerful and highly customizable engine. Finally. 1. let’s have a look at the “forest” from the top . a black box solution to your needs is inadequate. To hopefully address this issue. these chapters read a bit more like a reference manual. If you’re instead an intermediate user of Dynare. the User Guide goes into some depth in covering the theoretical underpinnings and methodologies that Dynare follows to solve and estimate DSGE models. At that point. These chapters can also serve as a basic primer if you are new to the practice of DSGE model solving and Bayesian estimation. simulate and estimate DSGE models. with an intuitive front-end interface. These are available in the “behind the scenes of Dynare” chapters 7 and 8.

WHAT IS DYNARE? 3 Figure 1.1: The . but for now it may be helpful to summarize what Dynare is able to do: . the model and its related attributes. This not only facilitates the inputting of a model. which then calls the relevant Matlab routines to carry out the desired operations and display the results.mod file being read by the Dynare pre-processor.mod file into a suitable input for the Matlab routines (more precisely. Each of these steps will become clear as you read through the User Guide. results are presented in Matlab.2. Figure 1.2 in chapter 4. but also enables you to easily share your code as it is straightforward to read by anyone. This initiates the Dynare pre-processor which translates the . In slightly less flowery words. it creates intermediary Matlab or C files which are then used by Matlab code) used to either solve or estimate the model. like a shock structure for instance. The resulting file will be called the . Some more details on the internal files generated by Dynare is given in section 4. it is a pre-processor and a collection of Matlab routines that has the great advantages of reading DSGE model equations written almost as in an academic paper. is written equation by equation in an editor of your choice. Finally. Basically. That file is then called from Matlab.mod file.1.2 gives you an overview of the way Dynare works.

giving a clear definition and explanation of usage for each. The User Guide will often introduce you to a command in a rather loose manner (mainly through examples). run my members of the Dynare team. as you learn about new features.of .usually well documented . . workshops and seminars that may be of interest. • Frequently Asked Questions (FAQ): this section of the Dynare site emphasizes a few of the most popular questions in the forums.3 Additional sources of help While this User Guide tries to be as complete and thorough as possible. Besides allowing you to stay up to date with the most recent papers and possibly make new contacts. it conveniently lists conferences. you have the following additional sources of help: • Reference Manual: this manual covers all Dynare commands. you will certainly want to browse other material for help. • Dynare forums: this lively online discussion forum allows you to ask your questions openly and read threads from others who might have run into similar difficulties. • DSGE. • Official online examples: the Dynare website includes other examples . so reading corresponding command descriptions in the Reference Manual is a good idea to cover all relevant details. is a resource for all scholars working in the field of DSGE modeling.mod files covering models and methodologies introduced in recent papers. At your disposal.4 • compute the steady state of a model CHAPTER 1.net: this website. and yearn to ask that one question whose answer seems to exist no-where. INTRODUCTION • compute the solution of deterministic models • compute the first and second order approximation to solutions of stochastic models • estimate parameters of DSGE models using either a maximum likelihood or a Bayesian approach • compute optimal policies in linear-quadratic models 1. struggle with adapting examples to your own work.

g. are forbidden. If. in the Newton algorithm for deterministic models and in the linearizations necessary to solve stochastic models). and spaces. • Double indicates a double precision number.is version 4. The drawback is that Dynare can now handle only a limited set of functions.for which this guide is written . as well as improvements. • Command is an instruction to Dynare or other program when specified. 1. NOMENCLATURE 5 1. Note that Matlab requires that names of files or functions start with alphabetical characters.1d3. 1. With respect to version 3. The major new features are the following: • Analytical derivatives are now used everywhere (for instance. . this concerns your Dynare . for instance.1e3. what’s new and backward compatibility The current version of Dynare . NOTE! These must start with an alphabetical character and can only contain other alphabetical characters and digits. 1.4 Nomenclature To end this introduction and avoid confusion in what follows.1. • Typewritten text indicates text as it should appear in Dynare code. the option must be specified in parenthesis in Dynare.1E-3. although in nearly all economic applications this should not be a constraint.1E3. All other characters.5 v4.mod files. • Integer indicates an integer number. • Parameter name indicates a parameter name which must follow the same naming conventions as above. as well as underscores ( ). The following syntaxes are valid: 1.4. including accents. This increases computational speed significantly. this new version introduces several important features. • Options or optional arguments for a command are listed in square brackets [ ] unless otherwise noted. Many of these are shared with the Reference Manual. it is worthwhile to agree on a few definitions of terms. it will show up in the Guide as [(option)].1D3 • Expression indicates a mathematical expression valid in the underlying language (e. 1. 1. • Variable name indicates a variable name. optimizations of routines and bug fixes. Matlab). • Filename indicates a file name valid in your operating system.

• Speed. Recall that in version 3. INTRODUCTION • Variables and parameters are now kept in the order in which they are declared whenever displayed and when used internally by Dynare. The names of the files internally generated by Dynare have also changed. (** more on this when explaining internal file structure TBD) • The syntax for the external steady state file has changed. This should be most noticeable when solving deterministic models. Several large-scale improvements have been implemented to speed up Dynare. in section 3.6 CHAPTER 1.3. This is covered in more details in chapter 3. .6. variables and parameters where at times in their order of declaration and at times in alphabetical order. but also apparent in other functionality. • The names of many internal variables and the organization of output variables has changed. NOTE! This may cause some problems of backward compatibility if you wrote programs to run off Dynare v3 output. NOTE! You will unfortunately have to slightly amend any old steady state files you may have written. These are enumerated in details in the relevant chapters.

Chapter 2 Installing Dynare 2. as well as Unix-like operating systems. 2. like the very processor intensive Metropolis Hastings algorithm. 2. you may need up to 1GB of RAM to obtain acceptable computational times. Note that Dynare++ is distributed along with Dynare since version 4. Development of the Scilab version stopped after Dynare version 3. You may also be interested by another program.1.2. although 512MB is preferred.1) runs on both MATLAB and GNU Octave.1 Dynare versions The current version of Dynare (4. Dynare++. 7 .02 and that for Gauss after Dynare version 1. in particular GNU/Linux and Mac OS X. If you have questions about the support of a particular platform. please ask your question on Dynare forums. Depending on the type of computations required.0 and later. There used to be versions of Dynare for Scilab and Gauss.3 Installing Dynare Please refer to the section entitled “Installation and configuration” in the Dynare reference manual.2 System requirements Dynare can run on Microsoft Windows. To run Dynare. which is a standalone C++ program specialized in computing k-order approximations of dynamic stochastic general equilibrium models. it is recommended to allocate at least 256MB of RAM to the platform running Dynare. This User Guide will exclusively focus on Dynare version 4. See the Dynare++ webpage for more information.

or to ask your particular question). INSTALLING DYNARE 2.8 CHAPTER 2. but even then remedies exist (see the Dynare forums for discussions on this. both of which are defined later. you will have additional options for solving for the steady state (solve algo option) and for searching for the posterior mode (mode compute option). no additional toolbox is necessary for running most of Dynare. But if you do have the ‘optimization toolbox’ installed. except maybe for optimal simple rules (see chapter 9). .4 MATLAB particularities A question often comes up: what special MATLAB toolboxes are necessary to run Dynare? In fact.

This chapter covers all these topics. agents will take their decisions 9 .this will serve as a base to carry out any of the above computations.is notable and writing two chapters would have been overly repetitive.basics This chapter covers everything that leads to. Typically. but you may also want to go a bit further. here is the important question: is your model stochastic or deterministic? The distinction hinges on whether future shocks are known. only the distribution of future shocks is known. In a stochastic context.Dynare commands and syntax . Let’s consider a shock to a model’s innovation only in period 1. This distinction will appear throughout the chapter. it is so fundamental. whether temporary or permanent.1 A fundamental distinction Before speaking of Dynare. the solution of DSGE models. You may be interested in simply finding the solution functions to a set of first order conditions stemming from your model. to learn basic Dynare commands and the process of writing a proper . you may be interested in how this system behaves in response to shocks. agents will take their decisions knowing that future values of the innovations will be zero in all periods to come. in fact.mod file . In stochastic models. In a deterministic context. That is to say that the term “solution” in the title of the chapter is used rather broadly. and stems from. instead. In deterministic models. it is important to recognize a distinction in model types. we recommend that you read this chapter chronologically.Chapter 3 Solving DSGE models . that we considered writing separate chapters altogether. a vast terrain. Likewise. Enough suspense. But the amount of common material . you may want to explore how the system comes back to its steady state or moves to a new one. the occurrence of all future shocks is known exactly at the time of computing the model’s solution. But instead of skipping to the topic closest to your needs. 3.

They can also last one or several periods. if an agent has perfect foresight. but also because it has been a source of significant confusion in the past. The solution is nothing more than a series of numbers that match a given set of equations. “stochastic”. perfect foresight and no uncertainty around shocks. not deterministic!) literature has gained attention in economics. This isn’t the same thing because of Jensen’s inequality. In this case. the following gives some additional details. for instance.1 NOTE! Deterministic vs stochastic models Deterministic models have the following characteristics: 1. 3. in which case they would be expected with perfect foresight. Most often. Shocks can hit the economy today or at any time in the future. policy or feedback rule for the future: what will her optimal actions be contingent on each possible realization of shocks. 4. if you consider only a first order linear approximation of the stochastic model. These models are usually introduced to study the impact of a change in regime. In deterministic models. a highly accurate solution can be found by numerical methods. 5. models introduce a positive shock today and zero shocks thereafter (with certainty). solutions to deterministic models are usually called “closed loop” solutions. Intuitively. or a linear model. . i. 3. and those to stochastic models are referred to as “open loop”. As the DSGE (read. deterministic models have become somewhat rare. she can specify today .1. as in the introduction of a new tax. SOLVING DSGE MODELS .10 CHAPTER 3. Models assume full information.what each of her precise actions will be in the future.e. The solution method for each of these model types differs significantly. Of course. Examples include OLG models without aggregate uncertainty. we therefore search for a function satisfying the model’s first order conditions. In a stochastic environment. In control theory.at the time of making her decision . this function may be non-linear and thus needs to be approximated. Because this distinction will resurface again and again throughout the chapter. 2. though. A second order approximation will instead lead to very different results. To complicate things. the best the agent can do is specify a decision.BASICS knowing that the future value of innovations are random but will have zero mean. instead. the two cases become practically the same. due to certainty equivalence. as the variance of shocks will matter.

Examples include most RBC models. which is the certainty equivalence property. 7. Stochastic models. 3. it involves numerical simulation to find the exact paths of endogenous variables that meet the model’s first order conditions and shock structure. The solution does not require linearization. have the following characteristics: 1.3. The model introduced here is a basic RBC model with monopolistic competition. More complex examples are instead presented in the advanced chapters. We will none-the-less illustrate both the stochastic and the deterministic settings on the basis of this example. or permanent changes in the exogenous variables cannot be handled due to the use of Taylor approximations around a steady state. are kept as bare as possible. Instead. when thinking of the latter. shocks hit today (with a surprise). instead. Note that as a general rule. 2. this is a good place to look for additional information on any of the following model set-up and discussion. in fact. as per the earlier discussion. Future sections will aim to code this example into Dynare and analyze its salient features under the influence of shocks .both in a stochastic and a deterministic environment. These types of models tend to be more popular in the literature. with just enough features to help illustrate Dynare commands and functionalities. Thus.2. Its particular notation adopted below is drawn mostly from notes available on Jesus Fernandez-Villaverde’s very instructive website. 3 and 5. or new keynesian monetary models.2 Introducing an example The goal of this first section is to introduce a simple example. Expected future shocks. used widely in the literature. This is an often overlooked point in the literature which misleads readers in supposing their models may be deterministic. This solution method can therefore be useful when the economy is far away from steady state (when linearization offers a poor approximation). Note throughout this model description that the use of expectation signs is really only relevant in a stochastic setting. 3. but thereafter their expected value is zero. Note that when these models are linearized to the first order. In these models. you’ll have to use a bit of imagination (on top of that needed to think you have perfect foresight!) to . the examples in the basic chapters. agents behave as if future shocks where equal to zero (since their expectation is null). INTRODUCING AN EXAMPLE 11 6. it doesn’t even really need a steady state.

These equation are 1 1 = βEt (1 + rt+1 − δ) ct ct+1 and ψ ct = wt 1 − lt The firm side of the problem is slightly more involved. we obtain the intuitive result that it = kt+1 − (1 − δ)kt . the more leisure due to the decreasing marginal utility of consumption). leisure and capital stock. The above equation can be seen as an accounting identity. As a consequence. 1−lt . with total expenditures on the left hand side and revenues .on the right hand side. production depends on capital). ∀t > 0 where kt is capital stock. In any given period. as the derivations are relatively standard. due to monopolistic competition. . the consumer therefore faces a tradeoff between consuming and investing in order to increase the capital stock and consuming more in following periods (as we will see later. total payments to factors. where lt is labor input. equals yt .BASICS ignore the expectation signs. yields the Euler equation in consumption.12 CHAPTER 3. Maximization of the household problem with respect to consumption. or that investment replenishes the capital stock thereby countering the effects of depreciation. with a little hand-waiving involved. Households maximize utility over consumption. according to the following utility function ∞ Et t=0 β [log ct + ψ log(1 − lt )] and subject to the following budget constraint ct + kt+1 = wt lt + rt kt + (1 − δ)kt .including the liquidation value of the capital stock . the equation can also be interpreted as a capital accumulation equation after bringing ct to the right hand side and noticing that wt lt + rt kt . rt real interest rates or cost of capital and δ the depreciation rate. SOLVING DSGE MODELS . or aggregate output. capturing the intertemporal tradeoff mentioned above. ct and leisure. wt real wages. and the labor supply equation linking labor positively to wages and negatively to consumption (the wealthier. if we define investment as it = yt − ct . with a little more imagination. Alternatively. but is presented below in the simplest possible terms. by the zero profit condition.

instead. Production of intermediate goods follows a CRS production function defined as α yit = kit (ezt lit )1−α where the i subscript stands for firm i of a continuum of firms between zero and one and where α is the capital elasticity in the production function. The solution to the sourcing problem yields an optimal capital to labor ratio. face a two pronged decision: how much labor and capital to employ given these factors’ perfectly competitive prices and how to price the variety they produce. Also. we combine the optimal capital to labor ratio into the production function and take advantage of its CRS property to solve for the amount of labor or capital required to produce one unit of output. instead. zt captures technology which evolves according to zt = ρzt−1 + et where ρ is a parameter capturing the persistence of technological progress and et ∼ N (0. The real cost of using this amount of any one . with 0 < α < 1. If we follow the second route. Intermediate producers. the final goods producer chooses his or her optimal demand for each variety. yielding the Dixit-Stiglitz downward sloping demand curve.2. we therefore have: mct = ( − 1)/ But what are marginal costs equal to? To find the answer. yields the well-known constant markup pricing condition of monopolistic competition: pit = −1 mct pt where pit is firm i’s specific price. We can either assume that firms sell differentiated varieties of a good to consumers who aggregate these according to a CES index. An additional step simplifies this expression: symmetric firms implies that all firms charge the same price and thus pit = pt . mct is real marginal cost and pt is the aggregate CES price or average price. Or we can postulate that there is a continuum of intermediate producers with market power who each sell a different variety to a competitive final goods producer whose production function is a CES aggregate of intermediate varieties. INTRODUCING AN EXAMPLE 13 There are two ways to introduce monopolistic competition. σ).3. or relationship between payments to factors: kit rt = α wt lit 1−α The solution to the pricing problem.

to close. we obtain mct = 1 1−α 1−α 1 α α 1 1−α α w rt At t which does not depend on i. which is the definition it of marginal cost: the cost in terms of labor input of producing an additional unit of output.we obtain aggregate production rt = α α 1−α yt = At kt lt which can be shown is equal to the aggregate amount of varieties bought by the final good producer (according to a CES aggregation index) and. Note. which is the same for all firms and thus does not depend on i. SOLVING DSGE MODELS . we aggregate the production of each individual firm to find an aggregate production function. with the optimal pricing condition yields the final two important equations of our model wt = (1 − α) and yit ( − 1) lit yit ( − 1) kit To end. let’s roll up our sleeves and see how we can input the model into Dynare and actually test how the model will respond to shocks. kt /lt . that because the ratio of output to each factor is the same for each intermediate firm and that firm specific as well as aggregate production is CRS. by using the optimal capital ∂l to labor ratio. On the other side. When solving for labor. as hinted earlier. the above can be worked out. Interestingly. we can rewrite the above two equations for wt and rt without the i subscripts on the right hand side. it is thus the same for all firms. for instance. or wt ∂yit . we factor out the capital to labor ratio. and noting that price dispersion is null . we have the Dixit-Stiglitz demand for each variety. This ends the exposition of the example.14 CHAPTER 3. equal to the aggregate output of final good. On the supply side. itself equal to household consumption. This should not be a surprise since the optimal capital to labor ratio follows from the maximization of the production function (minus real costs) with respect to its factors. . pit = pt . By equating the two and integrating both side. as well as its counterpart in terms of capital. in turn. to yield wt [(1 − α)yit /lit ]−1 . Now.or that.BASICS factor is given by wt lit + rt kit where we substitute out the payments to the other factor using again the optimal capital to labor ratio. Combining this result for marginal cost.

as mentioned loosely in the introduction of this Guide.mod file as containing four distinct blocks. DYNARE .mod file. illustrated in figure 3.g.mod file.MOD FILE STRUCTURE 15 3. external or internal to Matlab. The commands are: • var starts the list of endogenous variables. But before we get into executing a . forecasting. or the starting point for simulations or impulse response functions based on the model’s solution.3 Dynare . It will then be read by Matlab by first navigating within Matlab to the directory where the .mod file structure Input into Dynare involves the . (although actually typing the extension . to be separated by commas.mod file is stored and then by typing in the Matlab command line Dynare filename.3: • preamble: lists variables and parameters • model: spells out the model • steady state or initial value: gives indications to find the steady state of a model.3.4 Filling out the preamble The preamble generally involves three commands that tell Dynare what are the model’s variables. which are endogenous and what are the parameters.mod file can be written in any editor. 3. . let’s differentiate between the stochastic and deterministic cases. we lay these out. • varexo starts the list of exogenous variables that will be shocked. let’s start by writing one! It is convenient to think of the . First. then we discuss them.mod is not necessary). • shocks: defines the shocks to the system • computation: instructs Dynare to undertake specific operations (e.mod. In the case of our example. estimating impulse response functions) Our exposition below will structured according to each of these blocks. • parameters starts the list of parameters and assigns values to each. The .3.

et . beta = 0.16 CHAPTER 3. we go back to considering the law of motion for technology.4.mod file contains five logically distinct parts. the preamble would look like: var y c k i l y l w r.023.75. With respect to the above.1 The deterministic case The model is inherited exactly as specified in the earlier description. parameters beta psi delta alpha sigma epsilon.2 The stochastic case In this case. except that we no longer need the et variable. consisting of an exogenous shock. Thus. SOLVING DSGE MODELS . 3.007/(1-alpha)).33.99.BASICS Figure 3. varexo z. as we can make zt directly exogenous. 3. epsilon = 10.4. psi = 1. we therefore . delta = 0. alpha = 0.1: The . sigma = (0.

3. beta = 0. TIP! You can also comment out any line by starting the line with two forward slashes (//). writing a . rho = 0. beta = 0. although a single instruction can span two lines if you need extra space (just don’t put a semicolon at the end of the first line).75.75. . // the above instruction reads over two lines /* the following section lists several parameters which were calibrated by my co-author. Ask her all the difficult questions! */ alpha = 0. psi = 1. Two quick comments: NOTE! Remember that each instruction of the .99. psi = 1.mod file is really quite straightforward.33. varexo e. For example: var y c k i l y l w r z.3. alpha = 0.33. sigma = (0. delta = 0. Here’s what the preamble would look like: var y c k i l y l w r z.007/(1-alpha)).3 Comments on your first lines of Dynare code As you can tell.). rho = 0. parameters beta psi delta alpha rho sigma epsilon.4.4. or comment out an entire section by starting the section with /* and ending with */. epsilon = 10. delta = 0.mod file must be terminated by a semicolon (. and add the parameter ρ. FILLING OUT THE PREAMBLE 17 adjust the list of endogenous and exogenous variables.99.023. varexo e.95.95. parameters beta psi delta alpha rho sigma epsilon.023.

007/(1-alpha)). NOTE! that the above model specification corresponds to the stochastic case. SOLVING DSGE MODELS . This greatly facilitates the sharing of your Dynare files.BASICS sigma = (0.5. The fifth and sixth are the marginal cost equal to markup equations. as per our discussion of the preamble. The seventh is the investment equality. Let’s first have a look at our model in Dynare notation. The second the labor supply function. here’s a brief description: the first equation is the Euler equation in consumption. The eighth an identity that may be useful and the last the equation of motion of technology. The corresponding model for the deterministic casce would simply loose the last equation. epsilon = 10. 3.5 3. . c+i = y. What you can already try to do is glance at the model block below and see if you can recognize the equations from the earlier example. There are just a few conventions to follow.1 Specifying the model Model in Dynare notation One of the beauties of Dynare is that you can input your model’s equations naturally. as your colleagues will be able to understand your code in no-time. w = y*((epsilon-1)/epsilon)*(1-alpha)/l. y l = y/l. See how easy it is to read Dynare code? model. The third the accounting identity. psi*c/(1-l) = w. r = y*((epsilon-1)/epsilon)*alpha/k(-1). y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha). (1/c) = beta*(1/c(+1))*(1+r(+1)-delta).18 CHAPTER 3. end. The fourth is the production function. notice that the law of motion for technology is included. z = rho*z(-1)+e. Just in case you need a hint or two to recognize these equations. almost as if you were writing them in an academic paper. and then go through the various Dynare input conventions. indeed. i = k-(1-delta)*k(-1).

). each line of instruction ends with a semicolon (except when a line is too long and you want to break it across two lines. For instance. For example. but yesterday (recall that it is a function of yesterday’s investment and capital stock). 3. no matrix representation is necessary. . as in the preamble and everywhere along the .3 Notational conventions • Variables entering the system with a time t subscript are written plainly. . • Fourth. TIP! remember that variable and parameter names are case sensitive. xt+2 would be written x(+2).5. • Variables entering the system with a time t − n subscript are written with (−n) following them. • The first thing to notice. • Third. For example. more on this below . there need to be as many equations as you declared endogenous variables (this is actually one of the first things that Dynare checks. 3. . we would translate this equation into Dynare as k=i+(1-delta)*k(-1).5.2 General conventions The above example illustrates the use of a few important commands and conventions to translate a model into a Dynare-readable .4 Timing conventions • In Dynare. is that the model block of the . . xt would be written x. the timing of each variable reflects when that variable is decided. xt−2 would be written x(−2) (incidentally. in between. it will immediately let you know if there are any problems). but this notation makes it slightly harder to count by hand the number of forward looking variables (a useful measure to check). our capital stock is not decided today. Writing x(2) is also allowed. This is unlike Matlab where if you break a line you need to add .mod file begins with the command model and ends with the command end.mod file. SPECIFYING THE MODEL 19 3. Note that variable and parameter names used in the model block must be the same as those declared in the preamble. variables entering the system with a time t+n subscript are written with (+n) following them. For example.5. • Second. Thus.5. it is what we call in the jargon a predetermined variable. .3. eventhough in the example presented above we wrote kt+1 = it + (1 − δ)kt . as in many papers. equations are entered one after the other. this would count as two backward looking variables). • In the same way.mod file.

5. you must use a “stock at the end of the period” concept. If this condition is not met. would look like: model (linear).6 Linear and log-linearized models There are two other variants of the system’s equations which Dynare accommodates. • Blanchard-Kahn conditions are met only if the number of non-predetermined variables equals the number of eigenvalues greater than one. the model in exp-logs. in the equation for wages.5. Thus.BASICS • As another example. all that is necessary is to write the term (linear) next to the command model. you can write wage in period t (when they are set). Be careful.ll. this turns out to be a very useful option . 3. one needing to be greater and the other smaller than one for stability. First. with just the equation for yl for illustration. wages should appear with a one period lag. Our example. consumption could appear with a lead in the Euler equation. It is investment during period t that sets stock at the end of period t. 3. SOLVING DSGE MODELS . wages used during a period are set the period before. In the first case. • A slightly more roundabout way to explain the same thing is that for stock variables.20 CHAPTER 3. the second order difference equation would have two eigenvalues. For instance. you may be interested to have Dynare take Taylor series expansions in logs rather than in levels. where repeating a letter for a variable means difference from steady state. In this case. yy l=yy . Dynare will put up a warning. but also with a lag in a habit formation equation. • Note that a variable may occur both as predetermined and non-predetermined. consider that in some wage negociation models. Otherwise. if you had one. the linear model and second. end. a lot of papers use the “stock at the beginning of the period” convention. but in the labor demand equation.5 Conventions specifying non-predetermined variables • A (+1) next to a variable tells Dynare to count the occurrence of that variable as a jumper or forward-looking or non-predetermined variable. as we did (on purpose to highlight this distinction!) in the setup of the example model above.

as well as tips to do so more efficiently.6. you may be interested to start your simulations or impulse response functions from either a steady state. or another given point. your model will need to be linearized before it is solved. psi*exp(cc)/(1-exp(ll)) = exp(ww). they need to have a steady state. This section is also useful to specify this starting value. where.3. irrespective of whether you’re working with a stochastic or deterministic model. (1/exp(cc)) = beta*(1/exp(cc(+1)))*(1+exp(rr(+1))-delta). SPECIFYING STEADY STATES AND/OR INITIAL VALUES 21 when estimating models with unit roots. more rarely. recall that stochastic models need to be linearized. or approximations of values. To do so. First.3 below). Second. this time. as in the following fashion: initval. histval which is covered only in the Reference Manual. you should get through unscathed. these values are entered in the initval block. though.1 Stochastic models and steady states In a stochastic setting.6. note that the relevant commands in this section are initval. In either case. You can either enter exact steady state values into your . or just approximations and let Dynare find the exact steady state (which it will do using numerical methods based on your approximations).6 Specifying steady states and/or initial values Material in this section has created much confusion in the past. But with some attention to the explanations below. Dynare needs to know your model’s steady state (more details on finding a steady state. In passing. The Dynare input convention makes this very easy to do.6.mod file. Let’s see in more details how all this works. repeating a letter for a variable means log of that variable. Thus. 3. Our example would need to be re-written as follows (just shown for the first two equations) model. One of the functions of this section is indeed to provide these steady state values. so that the level of a variable is given by exp(repeatedvariable). The first two are instead covered in what follows. end. are provided in section 3.mod file. If so. . endval or. as we will see in chapter 5. simply rewrite your equations by taking the exponential and logarithm of each variable. Let’s start by emphasizing the uses of this section of the . 3.

BASICS Then. l = 0. TIP! If you’re dealing with a stochastic model. z = 0. end. you can control whether you want to start your simulations or impulse response functions from the steady state. SOLVING DSGE MODELS .22 k = 9. z = 0. if you don’t add the command steady. w = 2. r = 0. r = 0. by using the command steady. CHAPTER 3. Adding steady just after your initval block will instruct Dynare to consider your initial values as mere approximations and start simulations or impulse response functions from the exact steady state.3. thus it is strongly recommended that you start your simulations from a steady state. e = 0. e = 0.7. c = 0.0. k = 9.3. your simulations or impulse response functions will start from your initial values. . steady. even if Dynare will have calculated your model’s exact steady state for the purpose of linearization.7. this means either using the command steady or entering exact steady state values. c = 0. On the contrary. w = 2. l = 0. remember that its linear approximation is good only in the vicinity of the steady state. end. the above block would be expanded to yield: initval. or from the exact values you specified in the initval block. For the case in which you would like simulations and impulse response functions to begin at the steady state.0.

their initial values would all be zero. it is better to start with a smaller model and add new variables one by one. Otherwise. the following TIPS! may help. Unfortunately. the initval block serves very similar functions as described above. if you wanted to begin your solution path from an arbitrary point.6. of course. Dynare can help in finding your model’s steady state by calling the appropriate Matlab functions. most researchers are still interested to see how a model reacts to shocks when originally in steady state. But practically. you may still run into difficulties in finding your steady state. If so. 3. Doing so borders on a form of art. and luck is unfortunately part of the equation. followed by the command steady. you can begin by playing with the options following the steady command.6. technically. you would enter those values in your initval block and not use the steady command. Yet. As mentioned above. This is the default option if none are specified. In this case. is calculating actual steady state values. If you have trouble finding the steady state of your model. variables would be expressed in percent deviations from steady state. Thus. another option is to enter your model in linear terms. you would enter approximate (or exact) steady state values in the initval block.6. For complicated models. Often. If you wanted to shock your model starting from a steady state value.3 Finding a steady state The difficulty in the above. SPECIFYING STEADY STATES AND/OR INITIAL VALUES 23 3. An illustration of the initval block in the deterministic case appears further below. Thus. your . In the deterministic case. But even for simpler models.2 Deterministic models and initial values Deterministic models do not need to be linearized in order to be solved. • solve algo = 3: uses the Sims solver.3. These are: • solve algo = 0: uses Matlab Optimization Toolbox FSOLVE • solve algo = 1: uses Dynare’s own nonlinear equation solver • solve algo = 2: splits the model into recursive blocks and solves each block in turn. finding suitable initial values for the endogenous variables is the trickiest part of finding the equilibrium of that model. if any of your original (non-linear) equations involve sums (a likely fact). you do not need to provide a steady state for these model. But it is usually only successful if the initial values you entered are close to the true steady state.

6. The alternative is to write a Matlab program to find your model’s steady state. you need to do a little work to write your steady state program. SOLVING DSGE MODELS .mod file and ask Matlab to solve the system. though (unless you’re working with a particular toolbox). But of course. in the stochastic case. As mentioned earlier. the steady state file corresponding to the above example. It is not enough to simply input the equations as you’ve written them in your . Alternatively. Because Matlab does not work with analytical expressions. If so. if your . this procedure could be time consuming and bothersome. you need to input your expressions sequentially. a necessary condition for the uniqueness of a stable equilibrium in the neighborhood of the steady state is that there are as many eigenvalues larger than one in modulus as there are forward looking variables in the system. becomes seamless.m) file should have the same name as your . non-linear. model. it will use that file to find steady state values regardless of whether you’ve provided initial values in your .mod file to see if such a Matlab file exists. Doing so has the clear advantages of being able to incorporate your Matlab program directly into your .mod file. That is.4 Checking system stability TIP! A handy command that you can add after the initval or endval block (following the steady command if you decide to add one) is the check command. would be: (** example file to be added shortly) 3. your matlab (. This computes and displays the eigenvalues of your system which are used in the solution method.m and should be saved in the same directory as your . NOTE! When doing so. You will instead need to write your steady state program as if you were solving for the steady state by hand. your Matlab file should be called example steadystate. Dynare will automatically check the directory where you’ve saved your . you may be left needing to calculate fewer steady state values than in the original. you could also use an external program to calculate exact steady state values.24 CHAPTER 3.mod file is called example.BASICS linearized equations will include ratios of steady state values. For instance. for instance. which you would still need to calculate.mod. you could write an external Maple file and then enter the steady state solution by hand in Dynare. especially if you want to alter parameter values (and thus steady states) to undertake robustness checks. whereby each left-hand side variable is written in terms of known parameters or variables already solved in the lines above.mod file. followed by steadystate For instance. For example. If this condition is not .mod file. Yet.mod file so that running loops with different parameter values.

2 Deterministic models .1. this would have corresponded to a 10% shock. such as periods 5:10. but would simply tell the system to which (steady state) values you would like it to move and let Dynare calculate the transition path. Given the above instructions.1 Adding shocks Deterministic models . The distinction is that under a temporary shock. you would not specify actual “shocks”. the shocks are entirely expected. while under a permanent shock. Dynare would replace the value of zt specified in the initval block with the value of 0. The corresponding instructions would be: . you are free to set the duration and level of the shock. values 0. such as a structural change in your model. ADDING SHOCKS 25 met.3. In both cases.7 3.permanent shocks To study the effects of a permanent shock hitting the economy today. 3. To specify a shock that lasts 9 periods on zt .7. For instance. you would use the endval block following the usual initval block. as explained in our original discussion on stochastic and deterministic models. the model eventually comes back to steady state. To do so.1 entered above. If variables were in logs. To work with a temporary shock. for instance. end. you would write: shocks. you have the choice of introducing both temporary and permanent shocks. you may specify all values to remain common between the two blocks. Finally. note that we could have entered future periods in the shocks block. Dynare will tell you that the Blanchard-Kahn conditions are not satisfied (whether or not you insert the check command). except for the value of technology which you may presume changes permanently.7. in order to study the anticipatory behavior of agents in response to future shocks. the model reaches a new steady state. Note that you can also use the mshocks command which multiplies the initial value of an exogenous variable by the mshocks value. though. var z.7. 3.temporary shocks When working with a deterministic model. periods 1:9.

1. endval. z = 0. For instance. w = 2. l = 0.will take longer to reach their new steady state values. k = 9. c = 0. but only in period 10. the other variables . In the above example. we make use of the second steady since the actual terminal steady state values are bound to be somewhat different from those entered above.BASICS initval. Then you would follow the above endval block with: shocks. steady. when solved. In our example.1 in period 1 (tomorrow) and thereafter. This is unusual.26 CHAPTER 3. . SOLVING DSGE MODELS . and serves the same functionality as described earlier (namely. which are nothing but the initial values for all variables except for technology. But of course. In this case. TIP! If you instead wanted to study the effects of a permanent but future shock (anticipated as usual). end. you would have to add a shocks block after the endval block to “undo” the first several periods of the permanent shock. var z. the value of technology would move to 0.1. w = 2. suppose you wanted the value of technology to move to 0. you may impose on your system that it does not return to steady state. r = 0. your problem would become a so-called two boundary problem. z = 0. where steady can also be added to the endval block.3.0. If you do not use steady after endval. and the latter does not list exact steady state values. l = 0. of telling Dynare to start and/ or end at a steady state close to the values you entered.7.3. steady. k = 9.0. c = 0. which. requires that the path of your endogenous variables pass through the steady state closest to your endval values). r = 0.the endogenous variables .7. end.

we have written an instructive . which is itself an AR(1). In that case. 27 3. end. Furthermore. though. exactly as in the model we introduced from the outset.mod file in section 3. shocks can only hit the system today. or a depreciation of your currency. 3. that affects the model’s true exogenous variable.8. where σ is determined in the preamble block. as we did in the preamble of the . This can be particularly useful if you’re studying the effects of anticipated shocks in a stochastic model. we will distinguish between deterministic and stochastic models.8 Selecting a computation So far.7. we would write: shocks. it will be impulse response functions (IRFs) due to the external shocks. SELECTING A COMPUTATION periods 1:9.4. as the expectation of future shocks must be zero. A permanent shock cannot be accommodated due to the need to stationarize the model around a steady state.3.mod file. values 0. With that in mind. Supposing we wanted to add a shock with variance σ 2 . but what should Dynare do with it? What are we interested in? In most cases. For instance. Again. TIP! You can actually mix in deterministic shocks in stochastic models by using the commands varexo det and listing some shocks as lasting more than one period in the shocks block. we would declare zt as an endogenous variable and et as an exogenous variable. you may be interested in what happens to your monetary model if agents began expecting higher inflation. zt in our example. please see the Reference Manual. we can however make the effect of the shock propagate slowly throughout the economy by introducing a “latent shock variable” such as et in our example. end.3 Stochastic models Recall from our earlier description of stochastic models that shocks are only allowed to be temporary. For information on how to do so. Let’s see which are the appropriate commands to give to Dynare. var e = sigma^2. .

2 For stochastic models In the more common case of stochastic models. you must specify a large enough number of periods such that increasing it further doesn’t change the simulation for all practical purpose.28 CHAPTER 3. That said. This command instructs Dynare to compute a Taylor approximation of the decision and transition functions for the model (the equations listing current values of the endogenous variables of the model as a function of the previous state of the model and current shocks). SOLVING DSGE MODELS . To do so.8.TIP! If you linearize your model up to a first order. In the case of a temporary shock. impulse response functions and various descriptive statistics (moments. Note that unless you use the endval command. 3. for instance. the algorithm makes the simplifying assumption that the system is back to equilibrium after the specified number of periods. the variance decomposition is computed as in the VAR literature through a Cholesky decomposition of the covariance matrix of the exogenous variables. the trajectory will basicaly describe how the system gets back to equilibrium after being perturbed from the shocks you entered. note that future shocks will not have a significant impact For correlated shocks.BASICS 3. the variance decomposition depends upon the order of the variables in the varexo command. impulse response functions will be the result of actual Monte Carlo simulations of future shocks. This is because in second order linear equations. If you instead linearize to a second order. and repeat this procedure a multitude of times in order to draw out an average response.mod file.1 Impulse response functions are the expected future path of the endogenous variables conditional on a shock in period 1 of one standard deviation.1 For deterministic models In the deterministic case. Note that the command takes the option [ (periods=INTEGER) ] The command simul triggers the computation a numerical simulation of the trajectory of the model’s solution for the number of periods set in the option. The technique is instead to pull future shocks from their distribution and see how they impact your system. the command stoch simul is appropriate. correlation and autocorrelation coefficients). variance decomposition. impulse response functions are simply the algebraic forward iteration of your model’s policy or decision rule. all you need to do is add the command simul at the bottom of your . When the shocks are correlated. Thus. 1 .8. it uses a Newton method to solve simultaneously all the equations for every period (see Juillard (1996) for details). so that the effects of the shocks depend on the state of the system when the shocks hit. it is impossible to get algebraic average values of all future shocks and their impact. Thus. you will have cross terms involving the shocks.

. If you’re interested to peer a little further into what exactly is going on behind the scenes of Dynare’s computations. • irf = INTEGER: number of periods on which to compute the IRFs (default = 40). • nocorr: doesn’t print the correlation matrix (printing is the default). • nofunctions: doesn’t print the coefficients of the approximated solution (printing is the default). Dynare will instead report theoretical moments. For a complete list of options. given their mean of zero. • hp filter = INTEGER: uses HP filter with lambda = INTEGER before computing moments (default: no filter). • relative irf requests the computation of normalized IRFs in percentage of the standard error of each shock. • dr algo = 0 or 1: specifies the algorithm used for computing the quadratic approximation of the decision rules: 0 uses a pure perturbation approach as in Schmitt-Grohe and Uribe (2004) (default) and 1 moves the point around which the Taylor expansion is computed toward the means of the distribution as in Collard and Juillard (2001b). It may be necessary to increase it for highly autocorrelated processes (default = 512). • hp ngrid = INTEGER: number of points in the grid for the discreet Inverse Fast Fourier Transform used in the HP filter computation. For first order linearizations. have a look at Chapter 7. SELECTING A COMPUTATION 29 on your results. the return to steady state is asymptotic. Details on implementing this appear below.3. Options following the stoch simul command: • ar = INTEGER: Order of autocorrelation coefficients to compute and to print (default = 5). since they get averaged between each Monte Carlo trial and in the limit should sum to zero. TIP! thus you should make sure to specify sufficient periods in your IRFs such that you actually see your graphs return to steady state. Dynare will return the actual sample moments from the simulations. Here instead. • drop = INTEGER: number of points dropped at the beginning of simulation before computing the summary statistics (default = 100). please see the Reference Manual. Note that in the case of a second order approximation. we focus on the application of the command and reproduce below the most common options that can be added to stoch simul.8. Setting IRF=0. In both cases. suppresses the plotting of IRF’s.

• replic = INTEGER: number of simulated series used to compute the IRFs (default = 1 if order = 1.000001). Dynare will actually undertake Monte Carlo simulations to generate moments of your variables. are basically happy with all default options and want to carry out simulations over a good number of periods. results are bound to be slightly different each time you run your program. unless you’re working with a linear model in which case the order is automatically set to 1. . but a simulation only repeats the process once. TIP! If you do decide to fix the seed. SOLVING DSGE MODELS . • periods = INTEGER: specifies the number of periods to use in simulations (default = 0). • qz criterium = INTEGER or DOUBLE: value used to split stable from unstable eigenvalues in reordering the Generalized Schur decomposition used for solving 1st order problems (default = 1. except if you fix the seed for the random number generator. If you linearized to a second order. • noprint: cancel any printing. you should at least try to run your program without using simul seed. Otherwise a different sample is used for each run (default: seed not specified). just to check the robustness of your results. suppose we were interested in printing all the various measures of moments of our variables. and 50 otherwise). Because of the simulation. Going back to our good old example. usefull for loops. • simul seed = INTEGER or DOUBLE or (EXPRESSION): specifies a seed for the random number generator so as to obtain the same random sample at each run of the program. whereas impulse response functions run a multitude of Monte Carlo trials in order to get an average response of your system. want to see impulse response functions for all variables. • order = 1 or 2 : order of Taylor approximation (default = 2). We would then end our . in the way that both sample shocks from their distribution to see how the system reacts.30 CHAPTER 3.BASICS • nomoments: doesn’t print moments of the endogenous variables (printing them is the default).mod file with the following command: stoch simul(periods=2100). TIP! A simulation is similar to running impulse response functions with a model linearized to the second order.

76. You can find the corresponding files in the models folder under UserGuide in your installation of Dynare.03. beta = 0.95. here are the complete . alpha = 0.9. 3. c+i = y. initval.MOD FILE 31 3. delta = 0. y l = y/l.mod for deterministic models.007/(1-alpha)). psi*c/(1-l) = w. end. w = y*((epsilon-1)/epsilon)*(1-alpha)/l. .9. y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha).9 The complete . sigma = (0.75. and for the pleasure of seeing our work bear its fruits. (1/c) = beta*(1/c(+1))*(1+r(+1)-delta). rho = 0. psi = 1. THE COMPLETE .3. varexo e. r = y*((epsilon-1)/epsilon)*alpha/k(-1). The files are called RBC Monop JFV.1 The stochastic model var y c k i l y l w r z. k = 9. c = 0. epsilon = 10.023. z = 0. model. i = k-(1-delta)*k(-1). w = 2. l = 0. r = 0.07. end.99.3. parameters beta psi delta alpha rho gamma sigma epsilon.mod files corresponding to our example for the deterministic and stochastic case.mod file For completion’s sake. e = 0.mod for stochastic models and RBC Monop Det.33. z = rho*z(-1)+e.

y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha).0.99. delta = 0. epsilon = 10. model. . (1/c) = beta*(1/c(+1))*(1+r(+1)-delta). check. c = 0.BASICS steady. varexo z. sigma = (0. psi = 1.33. steady.9. 3.7.023. shocks. var e = sigma^2.75. l = 0. SOLVING DSGE MODELS . beta = 0. z = 0. r = 0. c+i = y. r = y*((epsilon-1)/epsilon)*alpha/k(-1). i = k-(1-delta)*k(-1). k = 9. alpha = 0. psi*c/(1-l) = w. initval. w = 2. w = y*((epsilon-1)/epsilon)*(1-alpha)/l.3.007/(1-alpha)). parameters beta psi delta alpha sigma epsilon. end.32 CHAPTER 3. end. y l = y/l. stoch simul(periods=2100). end.2 The deterministic model (case of temporary shock) var y c k i l y l w r .

1. end. jumpers.stochastic models The tabular results can be summarized as follows: 1. . values 0. var z. dynare ExSolStoch. you should get two forms of output .3. to execute your .mod files are stored.. navigate within Matlab to the directory where the example . FILE EXECUTION AND RESULTS check.mod file. 33 3.10. (** note. You can do this by clicking in the “current directory” window of Matlab. for instance.. Model summary: a count of the various variable types in your model (endogenous. and you should see a confirmation of the Blanchard-Kahn conditions if you used the command check in your .). simul(periods=2100).10 File execution and results To see this all come to life.mod file.mod file.10.mod file corresponding to the stochastic model is called RBC Monop JFV. which is conveniently installed by default in the Dynare “examples” directory (the . periods 1:9. As a result. all you need to do is place your cursor in the Matlab command window and type.1 Results . Let’s review these results.mod files should take at most 30 seconds. Once there. or typing the path directly in the top white field of Matlab. etc. 3. shocks.mod and that corresponding to the deterministic model is called RBC Monop Det.mod file. Running these . let’s run our .tabular in the Matlab command window and graphical in one or more pop-up windows. this may not be the case when testing the beta version of Matlab version 4) To run a . 2.mod). Eigenvalues should be displayed.

you will see some intermediate output: the errors at each iteration of the Newton solver used to estimate the solution to your model.10.mod file. show the actual impulse response functions for each of the endogenous variables.BASICS 3. i. as specified in the options of stoch simul. If not. yt−1 . either check that you have included enough periods in your simulations. A first order approximation of this function can be written as yt = y + gy yt−1 + gu ut . with ¯ ˆ yt = yt − y and y being the steadystate value of y. The graphical results. that your steady state actually exists and is stable. you should detrend your variables and rewrite your model in terms of those variables.e. TIP! If some variables do not return to their steady state. 6. Moments of simulated variables: up to the fourth moments. 7. if all goes well! Finally. presented in a table. 5. Correlation of simulated variables: these are the contemporaneous correlations. ut )] = 0 . yt .34 CHAPTER 3. yt = g(yt−1 . In other words. means finding an unkown function. If you entered steady. and where gx is the ˆ ¯ ¯ partial derivative of the g function with respect to variable x. you may want to try to increase the periods for the transition to the new steady state (the number of simulations .2 Results . instead. If you entered check. If so. SOLVING DSGE MODELS . the function g is a time recursive (approximated) representation of the model that can generate timeseries that will approximatively satisfy the rational expectation hypothesis contained in the original model. 4.deterministic models Automatically displayed results are much more scarce in the case of deterministic models. Autocorrelation of simulated variables: up to the fifth lag. the table “Policy and Transition function” contains the elements of gy and gu . or make sure that your model is stationary. ut ) that could be plugged into the original model and satisfy the implied restrictions (the first order conditions). given that they actually moved. Policy and transition functions: Solving the rational exectation model. In Dynare. your model will probably not converge. you will get a list of your steady state results. Et [f (yt+1 . These can be especially useful in visualizing the shape of the transition functions and the extent to which each variable is affected. eigenvalues will also be displayed and you should receive a statement that the rank condition has been satisfied. TIP! You should see these errors decrease upon each iteration. if not. Matrix of covariance of exogenous shocks: this should square with the values of the shock variances and co-variances you provided in the . Details on the policy and transition function can be found in Chapter 6. 3.

10.1. FILE EXECUTION AND RESULTS 35 periods). it does not mean that you cannot create these by hand from Matlab. . it may be a good idea to revise your equations. Of course. although Dynare does not display a rich set of statistics and graphs corresponding to the simulated output.3. But more often. saving and viewing your output.3 of chapter 4 on finding. To do so. you should start by looking at section 4.

.

in the Official Examples section of the Dynare website. and/ or feel comfortable with.1 Dynare features and functionality Other examples Other examples of . the earlier chapter 3 on the basics of solving DSGE models. along with helpful notes and explanations. if you have read.that you will probably find interesting or at least understandable. In particular. Jesus Fernandez-Villaverde has provided a series of RBC model variants (from the most basic to some including variable capacity utilization. The goal is to provide a brief explanation of the files that are created by Dynare to help you in troubleshooting or provide a starting point in case you actually want to customize the way Dynare works.not all related to each other . 4. indivisible labor and investment specific technological change). using loops. this chapter is divided into three sections. don’t forget to check occasionally the Dynare contributions and examples forum to see if any other user has posted an example that could help 37 . You can find these.1.Chapter 4 Solving DSGE models advanced topics This chapter is a collection of topics . To provide at least some consistency. but possibly also helpful for other work.1 4.mod files used to generate impulse response functions are available on the Dynare website. referring to external files and dealing with infinite eigenvalues. The third section of the chapter focusses on modeling tips optimized for Dynare. Also. The second section overviews some of the inner workings of Dynare. finding and saving your output. such as dealing with correlated shocks. The first section deals directly with features of Dynare.

2 Alternative. Your model may have two or more shocks. As is standard. This is to give you an alternative. the discount factor 0 < β < 1.ADVANCED TOPICS you in your work. and these may be correlated to each other. full-blown example to the one described in chapter 3. We assume output is produced according to a standard constant returns to scale technology of the form α yt = exp(at )kt h1−α t with α being the capital elasticity in the production function. The economy is a real economy. The model The model is a simplified standard RBC model taken from Collard and Juillard (2003) which served as the original User Guide for Dynare. to learn the workings of Dynare. as usual. the example provided is somewhat more complete than strictly necessary. and where at represents a stochastic technological shock (or Solow residual). SOLVING DSGE MODELS . The example below illustrates how you would introduce this into Dynare. with 0 < α < 1. where part of output can be consumed and part invested to form physical capital. complete example The following example aims to give you an alternative example to the one in chapter 3. . or maybe you would like to post an example there yourself? 4. the disutility of labor θ > 0 and the labor supply elasticity ψ ≥ 0. the law of motion of capital is given by kt+1 = exp(bt )it + (1 − δ)kt with 0 < δ < 1.38 CHAPTER 4. Actually. The economy consists of an infinitely living representative agent who values consumption ct and labor services ht according to the following utility function ∞ Et τ =t β τ −t h1+ψ log(ct ) − θ t 1+ψ where. A social planner maximizes this utility function subject to the resource constraint ct + it = yt where it is investment and yt output. where δ is physical depreciation and bt a shock affecting incorporated technological progress.1. Consumers are therefore also owners of the firms. It also aims to give you exposure to dealing with several correlated shocks.

ct θh1+ψ = (1 − α)yt t 1 = βEt exp(bt )ct exp(bt+1 )ct+1 exp(bt+1 )α yt+1 +1−δ kt+1 α 1−α yt = exp(at )kt ht kt+1 = exp(bt )it + (1 − δ)kt at = ρat−1 + τ bt−1 + t bt = τ at−1 + ρbt−1 + νt The . u = phi*0.4. . Note that the first equation captures the labor supply function and the second the intertemporal consumption Euler equation.mod file To “translate” the model into a language understandable by Dynare. That is at bt = ρ τ τ ρ at−1 bt−1 + t νt where |ρ + τ | < 1 and |ρ − τ | < 1 to ensure stationarity (we call ρ the coefficient of persistence and τ that of cross-persistence). we assume Et ( t ) = 0. end. we have two options (this was not discussed in the earlier chapter).probably quite similar to standard RBC models you have run into . var e. var u. We will assume that you’re comfortable with these and simply present the final .mod file below. var e. Et (νt ) = 0 and that the contemporaneous variance-covariance matrix of the innovations t and νt is given by σ2 ψσ σν and where corr( t νs ) = 0. Fist. Either write: shocks. we would follow the steps outlined in chapter 3. we specify a shock structure that allows for shocks to display persistence across time and correlation in the current period. . stderr 0.009.yields the following first order conditions (which are straightforward to reproduce in case you have doubts.009. note that to introduce shocks into Dynare.009. DYNARE FEATURES AND FUNCTIONALITY 39 Finally.1. stderr 0. This system . ) and equilibrium conditions drawn from the description above. corr( t s) ψσ σν 2 σν = 0 and corr(νt νs ) = 0 for all t = s. though.009*0. . Furthermore.

varexo e. model.95. a = rho*a(-1)+tau*b(-1) + e. a. b = tau*a(-1)+rho*b(-1) + u. tau = 0. initval. parameters beta. k = exp(b)*(y-c)+(1-delta)*k(-1). var y. theta = 2. theta. k. . you can also write: shocks.009*0. h.99.mod. end.40 CHAPTER 4. b. u = phi*0.009^2. here is the complete . rho. You can find the corresponding file in the models folder under UserGuide in your installation of Dynare. SOLVING DSGE MODELS . psi = 0. y = exp(a)*(k(-1)^alpha)*(h^(1-alpha)).80359242014163. delta = 0. Alternatively.36. var e. beta = 0. The file is called Alt Ex1.1. phi = 0. end.mod file corresponding to the above example.025. rho = 0. c*theta*h^(1+psi)=(1-alpha)*y. var u = 0.009^2. u.95. k = beta*(((exp(b)*c)/(exp(b(+1))*c(+1))) *(exp(b(+1))*alpha*y(+1)+(1-delta)*k)). c.025. c = 0. y = 1.08068253095672. tau.009. var e = 0. alpha. delta.ADVANCED TOPICS where the last line specifies the contemporaneous correlation between our two exogenous variables. alpha = 0. psi. So that you can gain experience by manipulating the entire model.

and columns correspond to the variables lagged 1. stderr 0. k = 5. e = 0. and so on. gamma y{1} represents variances. u = 0. Thus.009.29175631001732. Each row of these matrices will correspond to a variables in time t. end. By default. var e. u = phi*0. 41 4. then lagged 2 for the second matrix. b = 0.1. shocks. or using the workspace interface. • gamma y: the matrices of autocovariances. DYNARE FEATURES AND FUNCTIONALITY h = 0. var u.mod file): • steady state: the steady state of your variables • mean: the mean of your variables • var: the variance of your variables • autocorr: the various autocorrelation matrices of your variables. while gamma y{2} represents autocovariances where variables on each column are lagged by one period and so on.1.009. The last matrix (gamma y{7} in . saving and viewing your output Where is output stored? Most of the moments of interest are stored in global variable oo You can easily browse this global variable in Matlab by either calling it in the command line. stoch simul(periods=2100).009. Dynare will return autocovariances with a lag of 5. running down each column.4. for the first matrix. In global variable oo you will find the following (NOTE! variables will always appear in the order in which you declared them in the preamble block of your . stderr 0. end.3 Finding. var e.009*0. the matrix of autocorrelations that is automatically displayed in the results after running stoch simul has. the diagonal elements of each of the various autocorrelation matrices described here. a = 0.

for instance. The advantage of using Matlab.irfs comprising of vectors named endogenous variable exogenous variable. this is a bit of a workaround. namely of specifying shocks in an external file. In Matlab. you will find a global variable oo . you could simulate a deterministic model with the shocks saved from the estimation by specifying the source file for the shocks. Finally.1. to find your model’s steady state was clear with respect to Dynare version 3. reporting the values of the endogenous variables corresponding to the impulse response functions. SOLVING DSGE MODELS . Dynare will save all endogenous variables. .mat) command. if you want to run loops.mat. like y e.6 of chapter 3 when discussing steady states. maybe. using the shocks(shocks file = datafile. But of course. For most usage scenarios. in which case writing a Matlab program may be more handy). either to compute the steady state of your model. But you may also be interested in the second possibility described above. you should therefore do just as well to ask Dynare to compute your model’s steady state (except. Furthermore. 4. where each column captures the independent contribution of each shock to the variance of each variable. or when specifying shocks in an external file.ADVANCED TOPICS the default case) returns the variance decomposition. as the latter resorted to numerical approximations to find steady state values. say.42 CHAPTER 4. To save your simulated variables. But Dynare version 4 now uses the same analytical methods available in Matlab. as a result of the independent impulse of each exogenous shock. as described in chapter 5. to vary your parameter values. you can add the following command at the end of your . since you could also use the built-in commands in Dynare to generate impulse response functions from estimated shocks.mod file: dynasave (FILENAME) [variable names separated by commas] If no variable names are specified in the optional field. variables saved with the dynasave command can be retrieved by using the Matlab command load -mat FILENAME. for instance. to simulate a model based on shocks from a prior estimation. You could then retrieve the exogenous shocks from the oo file by saving them in a file called datafile. The former is described in section 3.4 Referring to external files You may find it convenient to refer to an external file. if you decide to run impulse response functions.

you may get a message that there is an error in a file with a new name. The model equations residuals are stored in a vector named “residuals”.5 Infinite eigenvalues If you use the command check in your . parameters) are contained in a “y” (resp.m: a matlab file containing the model equations and their derivatives (first. up to the needed order (depending on the computing needs). third) derivatives are in “g2” matrix (resp. Endogenous variables (resp. Second (resp. don’t worry if you get infinite eigenvalues .4.m: a matlab file containing several instructions. then the pre-processor creates the following files: • filename. the pre-processor will output a C file (with . notably the parameter initializations and the matlab calls corresponding to computing tasks • filename dynamic. . They have no detrimental influence on the solution algorithm. The model jacobian is put in “g1” matrix.2. “params”) vector. The “y” vector has as many entries as their are (variable. Using a compiled C file is supposed to give better computing performance in model simulation/estimation. If the “use dll” option has been specified in the model declaration. 4. or you may want to have a closer look at how Dynare actually solves your model . Dynare will report your system’s eigenvalues and tell you if these meet the Blanchard-Kahn conditions.out of curiosity or maybe to do some customization of your own. (iii) outputting of several files. The dynare pre-processors essentially does three successive tasks: (i) parsing of the mod file (it checks that the mod file is syntactically correct). exogenous variables.2 Files created by Dynare At times. with an index number depending on the declaration order.these are are firmly grounded in the theory of generalized eigenvalues. and its translation into internal machine representation (in particular.1. As far as Blanchard-Kahn conditions are concerned infinite eigenvalues are counted as explosive roots of modulus larger than one. second and maybe third). which are used from matlab. model equations are translated into expression trees). “x”. If the mod file is “filename.mod”. (ii) symbolic derivation of the model equations. FILES CREATED BY DYNARE 43 4. “g3”).c extension) rather than a matlab file. It is then compiled to create a library (DLL) file.mod file. lag) pairs in the declared model. At that point. You may therefore find it helpful to get a brief overview of the internal files that Dynare generates and the function of each one.

Replaced by a C file when “use dll” option is specified.3. Of course.e. If you expect to see a growing curve for a variable. such that you can linearize them around a steady state and return to steady state after a shock. For example.mod file. you can always add it back after the simulation of the stationary components of the variables. either by hand. you must first stationarize your model. Again. More generally. if yt is I(1). you can always write yt+1 as yt + dyt+1 .3. SOLVING DSGE MODELS . 4.3 Infinite sums Dealing with infinite sums is tricky in general.1 Modeling tips Stationarizing your model Models in Dynare must be stationary. Used to compute the steady state.m: a matlab file containing the stationarized version of the model (i. you are thinking about a growth model. to enter the term Et−1 yt . not just a variable.44 CHAPTER 4. Note that in a stationary model. you would need to apply the above manipulation to the entire equation. because of Jensen’s inequality. it is expected that variables will eventually go back to steady state after the initial shock. Thus. the trick is to use only stationary variables in t + 1. Note that. You can then reconstruct ex-post the non-stationary simulated variables after running impulse response functions. where dyt = yt − yt−1 . Same notations than the dynamic file. suppose your model included: ∞ β j xt+j = 0. as if yt were an equation.3 4. you need to know the value of dyt at the final equilibrium. it is easier to work with the stationarized version of such models.3. if you know the trend. or by letting Dynare do the work. with its jacobian. and needs particular care when working with Dynare. 4. j=0 .ADVANCED TOPICS • filename static. If you do have non-linear terms on which you want to take expectations in the past.2 Expectations taken in the past For instance. Because growth models are nonstationary. 4. The trick is to use a recursive representation of the sum. define st = Et [yt+1 ] and then use s(−1) in your . then linearize it. For deterministic models. where lagged variables are replaced by current variables). you cannot do this for terms that enter your equation in a non-linear fashion.

S2t = yt + γS2t+1 . which can be written as a recursive system of the form: S1t = xt + βS1t+1 . as illustrated in the following brief example. ` la Calvo for instance. is ∞ p∗ (i) = µ + (1 − βθ) t k=0 (βθ)k Et [mcn (i)] t+k where µ is the markup. The trouble.4. This is particularly helpful. exemplified by papers such as Clarida. which can also be written in the following recursive manner: ∞ ∞ ∞ St ≡ j=0 β xt+j = xt + j=1 j β xt+j = xt + β j=0 j β j xt+1+j ≡ xt + St+1 This formulation turns out to be useful in problems of the following form: ∞ ∞ β xt+j = pt j=0 j=0 j γ j yt+j . i represents a firm of the continuum between 0 and 1. is instead typical of the new Keya nesian monetary literature. for instance. S1 = pt S2. as ∞ ∞ p∗ − pt−1 = (1 − βθ) t k=0 (βθ)k Et [mct+k ] + k=0 (βθ)k Et [πt+k ] . and mct is marginal cost as described in the example in chapter 3. given that it will be able to reset its price only with probability 1 − θ each period. Finally. after some algebraic manipulations. is how to input this infinite sum into Dynare? It turns out that the Calvo price setting implies that the aggregate price follows the equation of motion pt = θpt−1 + (1 − θ)p∗ .3. The RBC model with monopolistic competition introduced in chapter 3 involved flexible prices. The extension with sticky prices. defined as: ∞ St ≡ j=0 β j xt+j . thus implying the t following inflation relationship πt = (1 − θ)(p∗ − pt−1 ). Gali. in a Calvo type setting. we can also t rewrite the optimal price setting equation. β is a discount factor. and Gertler (1999). The optimal price for a firm resetting its price in period t. of course. MODELING TIPS 45 Note that the above can also be written by using an auxiliary variable St .

+Et−k xt . we can go one step further and write the above as πt = βEt [πt+1 ] + λmct where λ ≡ (1−θ)(1−βθ) . such as: Et−1 xt +Et−2 xt +. . Suppose your model included the following sum: ∞ yt = j=0 Et−j xt where yt and xt are endogenous variables.ADVANCED TOPICS where mct+k = mct+k +µ is the deviation of the marginal cost from its natural rate. by writing the right hand side as the first term of the sum (with k = 0) plus the remainder of the sum.3. which is the recognizable inflation equation in the new θ Keynesian (or new Neoclassical) monetary literature. . In Dynare. as the index of the expectations changes with each element of the sum. as in the following example. . defined as the marginal cost when prices are perfectly flexible. but for convenience. a different approach than the one mentioned above is necessary. Mathematically. the best way to handle this is to write out the first k terms explicitly and enter each one in Dynare.46 CHAPTER 4. which can be written as the left hand side term scrolled forward one period and appropriately discounted. The trick now is to note that the above can be written recursively. 4.4 Infinite sums with changing timing of expectations When you are not able to write an infinite sum recursively. SOLVING DSGE MODELS . this yields: p∗ − pt−1 = (1 − βθ)mct+k + πt + β θEt [p∗ − pt ] t t+1 which has gotten rid of our infinite sum! That would be enough for Dynare.

that subject is instead covered in some depth in chapter 8. Suppose also that we thought our little RBC model did a good job of reproducing reality. featuring a non-stationary model. we did not want to introduce yet another example in this section. This chapter is therefore very practically-oriented and abstracts from the underlying computations that Dynare undertakes to estimate a model. 5.1 Introducing an example The example introduced in this chapter is particularly basic. This is for two reasons. Second. but not necessarily to duplicate a “real life scenario” you would face when writing a paper.Chapter 5 Estimating DSGE models basics As in the chapter 3. so that by the end of the chapter you should have the capacity to estimate a model of your own. while more advanced topics of practical appeal are discussed in chapter 6. Instead. 47 . this chapter is structured around an example. Suppose we had data on business cycle variations of output. Recall from chapter 3 that we are dealing with an RBC model with monopolistic competition. the goal of the example in this chapter is really to explain features in context. there’s enough new material to keep you busy. First. We could then use Bayesian methods to estimate the parameters of the model: α. β. the discount factor. the capital share of output. we thought it would be easiest to simply continue working with the example introduced in chapter 3 with which you are probably already quite familiar. The goal of this chapter is to lead you through the basic functionality in Dynare to estimate models using Bayesian techniques. you can always move on to chapter 6 where you will find a full-fledged replication of a recent academic paper. Once you feel comfortable with the content of this chapter. though.

The model’s variables would therefore be stationary and we can proceed without complications. r = y*((epsilon-1)/epsilon)*alpha/k(-1). parameters beta psi delta alpha rho epsilon. w = y*((epsilon-1)/epsilon)*(1-alpha)/l. The alternative scenario with non-stationary variables is more complicated and dealt with in chapter 6 (in the additional example).but the Bayesian estimation procedure would still run successfully. our model block would look exactly as in chater 3: model.48 CHAPTER 5. (1/c) = beta*(1/c(+1))*(1+r(+1)-delta). c+i = y. ρ. y l = y/l. z = rho*z(-1)+e. the condition for undertaking estimation is that there be at least as many shocks as there are observables (a less stringent condition than for maximum likelihood estimation). the depreciation rate. psi*c/(1-l) = w. . 5. In the stationary case.mod file with: var y c k i l y l w r z. i = k-(1-delta)*k(-1). the markup parameter.2 Declaring variables and parameters To input the above model into Dynare for estimation purposes. the degree of persistence in productivity. we must first declare the model’s variables in the preamble of the . the weight on leisure in the utility function. We thus begin the . varexo e. Let’s see how to go about doing this. 5. and .yielding posterior distributions identical to prior distributions .mod file. ψ.3 Declaring the model Suppose that the equation of motion of technology is a stationary AR(1) with an autoregressive parameter. Note that in Bayesian estimation. less than one. ρ.BASICS δ. This is done exactly as described in chapter 3 on solving DSGE models. ESTIMATING DSGE MODELS . It may be that this does not allow you to identify all your parameters . y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha).

a steady state must exist for the model and although Dynare can calculate it.4. you will significantly slow down the computation of the posterior mode.3 of chapter 3. This is just as explained in details and according to the same syntax outlined in chapter 3. It is much more efficient to write an external Matlab steady state file and let Dynare use that file to find the steady state of your model by algebraic procedure. it first linearizes it around a steady state. Dynare recalculates the steady state of the model at each iteration of the optimization routine (more on this later).6.mod file.6 Declaring priors Priors play an important role in Bayesian estimation and consequently deserve a central role in the specification of the . Dynare must know which variables are observable for the estimation procedure. the steady state block will look exactly the same. by providing approximate initial values and relying solely on the builtin Dynare algorithm to find the steady state (a numerical procedure). Dynare will end up spending 60 to 70% of the time recalculating steady states. DECLARING OBSERVABLE VARIABLES end. 5. In fact. Thus.4 Declaring observable variables This should not come as a surprise.5 Specifying the steady state Before Dynare estimates a model. as this chapter uses the same model as that outlined in chapter 3. 5. The general syntax to introduce priors in Dynare is the following: . For the moment. in Bayesian estimation. please refer to section 3.5. we write: varobs y. covering the initval. 49 5. as explained in section 5. Thus. Priors. in finding the posterior mode.7 below. NOTE! These variables must be available in the data file. are declared as a distribution. we must give it a hand by declaring approximate values for the steady state. based on the newest round of parameters available. TIP! During estimation. For more details on writing an external Matlab file to find your model’s steady state. steady and check commands.

by using the parameters command and its related syntax. you should declare them as such. p4 ) IG1 (µ. you therefore have to put two empty spaces for parameters µ and σ. Note also that if some parameters in a model are calibrated and are not to be estimated. where the following table defines each term more clearly PRIOR SHAPE NORMAL PDF GAMMA PDF BETA PDF INV GAMMA PDF UNIFORM PDF Corresponding distribution N (µ. σ. say α. TIP! Choosing the appropriate prior for your parameters is a tricky. as well as the syntax to declare priors for maximum likelihood estimation (not Bayesian). Then.BASICS estimated params. think about the shape of your prior distribution. as explained in chapter 3. Ask yourself. p4 ] where µ is the PRIOR MEAN. σ is the PRIOR STANDARD ERROR. see the Reference Manual. . σ) U (p3 . PRIOR SHAPE. For a more complete review of all possible options for declaring priors. σ) G2 (µ. since the uniform distribution only takes p3 and p4 as arguments. . on which side? You may also go one step further and build a distribution for each of your parameters in your mind. +∞) [p3 . PRIOR STANDARD ERROR [. PARAMETER NAME. p3 is the PRIOR 3rd PARAMETER (whose default is 0) and p4 is the PRIOR 4th PARAMETER (whose default is 1). σ. you would write alpha. It is worth spending time on your choice of priors and to test the robustness of your results to your priors. for instance. you will necessarily also find a probability of zero in your posterior distribution. Should it be bounded? Should it be opened on either or both sides? Remember also that if you specify a probability of zero over a certain domain in your prior. what is the probability that your parameter is bigger than a certain value. think about the domain of your prior over each parameter. end. ESTIMATING DSGE MODELS . PRIOR MEAN. For instance.50 CHAPTER 5. and then specify parameters p3 and p4. First.PRIOR 4th PARAMETER] . p4 ] R+ [p3 . TIP! When specifying a uniform distribution between 0 and 1 as a prior for a parameter. Some considerations may prove helpful. p3 . PRIOR 3rd PARAMETER] [. p3 ) B(µ. uniform pdf. 0. and repeat the exercise by incrementally decreasing that value.1. You can then pick the standard distribution that best fits . Should it be symmetric? Skewed? If so. p4 ) Range R [p3 . yet very important endeavor.

# sig = 1/bet. 0. end. Coming back to our basic example. another useful command to use is the estimated params init command which declares numerical initial values for the optimizer when these are different from the prior mean. 0.99. 0. 0. you may find it easier to define a prior over the discount factor. by adding a pound sign (#) at the beginning of the corresponding line.if the optimizer got stuck the first time around. beta pdf. β. than its inverse which often shows up in Euler equations.5. rather than the parameter itself.05. instead of describing here the shapes and properties of each standard distribution available in Dynare.and wanting to enter the posterior mode as initial values for the parameters instead of a prior.02. beta pdf. Thus you would write: model. you are instead encouraged to visualize these distributions yourself. psi. inv gamma pdf. 0. delta. end. This is especially useful when redoing an estimation .normal pdf. gamma pdf.02.75. rho. stderr e. estimated params. 0. In such a case. 0. 0. bet. end.025. we would write: estimated params. DECLARING PRIORS 51 your perceived distribution. beta. 0. alpha.6. For example. 0. The Reference Manual gives more details as to the exact syntax of this command.003. TIP! It may at times be desirable to estimate a transformation of a parameter appearing in the model.95.35. gamma pdf. . Finally. epsilon. as a Matlab expression. either in a statistics book or on the Web. TIP! Finally. inf.05. or needing a greater number of draws in the Metropolis-Hastings algorithm .01.0. c = sig*c(+1)*mpk. 0. beta pdf.002. it is possible to declare the parameter to be estimated in the parameters statement and to define the transformation at the top of the model section.003. beta pdf. 10. 1.1.

datafile = FILENAME: the datafile (a . and variable names would show up in the first cell of each column. all that is necessary is to add the command estimation at the end of the .mat file. nograph: no graphs should be plotted. Easy enough. Despite this low default value. This is useful if model variables are in deviations from steady state. after the command estimation). .BASICS 5. we list the most common and useful options. This improves the computation of between group variance of the parameter means. mh replic should be larger than 1200 (default = 20000) 8. first obs = INTEGER: the number of the first observation to be used (default = 1).90) 7. one of the key criteria to evaluate the efficiency of the Metropolis-Hastings to evaluate the posterior distribution. nobs = INTEGER: the number of observations to be used (default: all observations in the file) 3. In Excel files. But the real complexity comes from the options available for the command (to be entered in parentheses and sequentially. This is useful when running loops. no prefiltering). or an . separated by commas. a . it is advisable to work with a higher value. mh nblocks = INTEGER: number of parallel chains for Metropolis Hasting algorithm (default = 2). 2. Below. or instance.52 CHAPTER 5. to divide the observations into sub-periods.mod file. for instance. and encourage you to view the Reference Manual for a complete list.m file.7 Launching the estimation To ask Dynare to estimate a model. 4. and therefore have zero mean. prefilter = 1: the estimation procedure demeans the data (default=0. such as 5 or more. 6. More details on this subject appear in Chapter 6.xls file). 5. but vectors of observations need to be named with the same names as those in var obs. ESTIMATING DSGE MODELS . Note that observations do not need to show up in any order. for instance. mh replic = INTEGER: number of replication for Metropolis Hasting algorithm. 1. Demeaning the observations would also impose a zero mean on the observed variables. observations could be ordered in columns. conf sig = {INTEGER — DOUBLE }: the level for the confidence intervals reported in the results (default = 0. For the time being.

A clear indication of a problem would be that the mode is not at the trough (bottom of the minus) of the posterior distribution. 2: uses Lester Ingber’s Adaptive Simulated Annealing. Dynare stores the mode (xparam1) and the hessian (hh) in a file called MODEL NAME mode. this means that the first half of the draws from the Metropolis-Hastings are discarded). 0: the mode isn’t computed. 13. This is a particularly helpful option to speed up the estimation process if you have already undertaken initial estimations and have values of the posterior mode. while if it were too low. mh drop = DOUBLE: the fraction of initially generated parameter vectors to be dropped before using posterior simulations (default = 0. This is helpful to diagnose problems with the optimizer. your Metropolis-Hastings iterations would never visit the tails of a distribution. 4 (default): uses Chris Sim’s csminwel. LAUNCHING THE ESTIMATION 53 9. The idea here is to draw initial values from a stretched out distribution in order to maximize the chances of these values not being too close together. 12. When computing the mode. mode check: when mode check is set.5. mode compute=INTEGER: specifies the optimizer for the mode computation. which would defeat the purpose of running several blocks of MetropolisHastings chains. mode file=FILENAME: name of the file containing previous value for the mode. 11. an acceptance rate of 25% in the Metropolis. This option must be tuned to obtain. ideally.2 and 0. the literature has settled on a value of between 0. 14. The idea is not to reject or accept too often a candidate parameter. The default value is rarely satisfactory. the iterations would get stuck in a subspace of the parameter range.2). Note that the acceptance rate drops if you increase the scale used in the jumping distribution and vice a versa.7.5. mh init scale=DOUBLE: the scale to be used for drawing the initial value of the Metropolis-Hastings chain (default=2*mh jscale). mode file must be specified 1: uses Matlab fmincon (see the Reference Manual to set options for this command). . Dynare plots the minus of the posterior density for values around the computed mode for each estimated parameter in turn.Hastings algorithm (default = 0. 3: uses Matlab fminunc. mh jscale = DOUBLE: the scale to be used for the jumping distribution in MH algorithm. If the acceptance rate were too high.4. 10.

Dynare adds to previous Metropolis-Hastings simulations instead of starting from scratch. on the other hand. not smoothed variables.8 of chapter 3.if not already . To calculate one period ahead prediction errors. including the variance of the shocks. 21. generates an IRF. Repeating this process often enough generates a distribution of IRFs. Dynare pulls parameter and shock values from the corresponding estimated distributions and.in Dynare version 4. ESTIMATING DSGE MODELS . smoother: triggers the computation of the posterior distribution of smoothed endogenous variables and shocks. 20. ** will be implemented shortly . 16.54 CHAPTER 5. are built only based on knowing past information. moments varendo: triggers the computation of the posterior distribution of the theoretical moments of the endogenous variables as in stoch simul (the posterior distribution of the variance decomposition is also included). nodiagnostic: doesn’t compute the convergence diagnostics for MetropolisHastings (default: diagnostics are computed and displayed). 19. TIP! If you stop the estimation procedure after calculating the posterior mode. . or carry out maximum likelihood estimation. bayesian irf: triggers the computation of the posterior distribution of impulse response functions (IRFs). either see the Reference Manual or section 3. To build the posterior distribution of the IRFs. instead. See the note below on the difference between filtered and smoothed shocks. The length of the IRFs are controlled by the irf option. Smoothed shocks are a reconstruction of the values of unobserved shocks over the sample. for instance. Again. filtered vars: triggers the computation of the posterior distribution of filtered endogenous variables and shocks. Actually seeing if the various blocks of Metropolis-Hastings runs converge is a powerful and useful option to build confidence in your model estimation. only the corresponding parameter estimates will be used to generate the IRFs. ** will be implemented shortly . All options available for stoch simul can simply be added to the above options. 18. load mh file: when load mh file is declared. To view a list of these options.if not already . as specified in chapter 3 when discussing the options for stoch simul. the IRFs will use the parameters the posterior distributions. 17. If you instead carry out a full Metropolis-Hastings estimation. separated by commas. this is a useful option to speed up the process of estimation. using all the information contained in the sample of observations. for each set of draws. More details on these diagnostics are given in Chapter 6.in Dynare version 4. Filtered shocks.BASICS 15. you should use filtered.

Note that Dynare cannot forecast out of the posterior mode.mh jscale=0. You need to run Metropolis-Hastings iterations before being able to run forecasts on an estimated model. you would comment out or erase the commands related to estimation.mod file.mh drop=0. In this test file. .8. Details on model solution and simulation can be found in Chapter 3. You can find the corresponding file in the models folder under UserGuide in your installation of Dynare. add a shocks block.8. This ends our description of the . as observation do not exist for all variables. forecast = INTEGER: computes the posterior distribution of a forecast on INTEGER periods after the end of the sample used in estimation.MOD FILE 55 22. THE COMPLETE . Finally. running a forecast is very similar to an IRF. those necessary are reconstructed by sampling out of the posterior distribution of parameters.mh nblocks=2. Again.8 The complete . mode compute=6). Finally. mh replic=2000. coming back to our example.mod file. except that the forecast does not begin at a steady state.in Dynare version 4. 5. we could choose a standard option: estimation(datafile=simuldataRBC. Of course. You may therefore want to create a test version of your . The goal of undertaking a forecast is to see how the system returns to steady state from this starting point. The corresponding graph includes one confidence interval describing uncertainty due to parameters and one confidence interval describing uncertainty due to parameters and future shocks. here is the complete . remove the prior estimates for parameter values and replace them with actual parameter values in the preamble. TIP! Before launching estimation it is a good idea to make sure that your model is correctly declared. make sure you have steady and possibly check following the initval block if you do not have exact steady state values and run a simulation using stoch simul at the end of your .first obs=500.nobs=200.mod file To summarize and to get a complete perspective on our work so far. ** will be implemented shortly .mod file. that a steady state exists and that it can be simulated for at least one set of parameter values.mod.5. repeating this step often enough yields a posterior distribution of the forecast.mod file for the estimation of our very basic model. remove any non-stationary variables from your model. The file is called RBC Est.if not already .45. but simply at the point corresponding to the last set of observations. as in bayesian irf.

beta pdf. w = y*((epsilon-1)/epsilon)*(1-alpha)/l. 0.95. inf. r = y*((epsilon-1)/epsilon)*alpha/k(-1). rho. y l = y/l.45.mh jscale=0. beta. ESTIMATING DSGE MODELS . 0. beta pdf. z = rho*z(-1)+e.75. varobs y. . mh replic=2000. i = k-(1-delta)*k(-1). y = (k(-1)^alpha)*(exp(z)*l)^(1-alpha). estimated params. 0. 0.mh drop=0. mode compute=6). 0.8.002. gamma pdf. 0. 0. z = 0.05.mh nblocks=2. varexo e.first obs=500.02.003. delta. stderr e.7. initval. steady.BASICS var y c k i l y l w r z. psi*c/(1-l) = w.01. check. 0. k = 9. r = 0. end.56 CHAPTER 5. estimation(datafile=simuldataRBC.nobs=200. c = 0. model. 0.02. 1. alpha.35. parameters beta psi delta alpha rho epsilon. end. w = 2.0. gamma pdf. inv gamma pdf. l = 0.3.99. beta pdf.025. end. 0. epsilon. psi. beta pdf. (1/c) = beta*(1/c(+1))*(1+r(+1)-delta). 0. e = 0. c+i = y.003. 10.

2 Graphical results ** corresponding graphs will be reproduced below.9. posterior mode. probably erroneous when undertaking Bayesian estimation. Dynare returns both tabular and graphical output. as explained in more details in Chapter 6. 5.9.hopefully! The next set of results are for the numerical iterations necessary to find the posterior mode. Note the dummy values of 1 for the nonstationary variables Y obs and P obs.mod file.5.9 Interpreting output As in the case of model solution and simulation. . Dynare give the value of the objective function (the posterior Kernel) at the mode and displays two important table summarizing results from posterior maximization.mod file above. The figure returns a graphical representation of the priors for each parameter of interest. These results are followed by the eigenvalues of the system. The first figure comes up soon after launching Dynare as little computation is necessary to generate it. INTERPRETING OUTPUT 57 5. this is simply the limit case of the inverse gamma distribution.9. as opposed to standard maximum likelihood). It may be entirely possible that you get an infinite value for a standard deviation. standard deviation and t-stat of the mode (based on the assumption of a Normal. as well as the prior distribution and standard deviation (pstdev). The table of eigenvalues is completed with a statement about the Blanchard-Kahn condition being met . The first table summarizes results for parameter values. The improvement from one iteration to the next reaches zero. On the basis of the options entered in the example .1 Tabular results The first results to be displayed (and calculated from a chronological standpoint) are the steady state results. It is followed by a second table summarizing the same results for the shocks. Dynare will display the following results. It includes: prior means. 5. presented in the order in which the endogenous variables are declared at the beginning of the .

being constructed from an 80% confidence interval around the parameter mean.BASICS The second set of figures displays “MCMC univariate diagnostics”. If the results from one chain are sensible. “m2”. where MCMC stands for Monte Carlo Markov Chains. Dynare reports both the within and the between chains measures. towards which most of the computations undertaken by Dynare are directed: the posterior distribution. These allow you to make statements about your data other than simply concerning the mean and variance of the parameters. results within any of the however many iterations of Metropolis-Hastings simulation should be similar. except that they reflect an aggregate measure based on the eigenvalues of the variance-covariance matrix of each parameter. Another approach is to undertake a greater number of Metropolis-Hastings simulations. This is the idea of what the MCMC diagnostics track. on the posterior distribution. you may have a problem of poor priors. Note that the horizontal axis represents the number of Metropolis-Hastings iterations that have been undertaken. grey lines). you can tell that indeed. corresponding to the measure at the initial value of the Metropolis-Hastings iterations. each time starting from a different initial value). This is the main source of feedback to gain confidence. Dynare reports three measures: “interval”. the figure compares the posterior to the prior distribution (black vs. The first to last figure . It is advisable to redo the estimation with different priors. The figure entitled “multivariate diagnostic” presents results of the same nature. Recall that Dynare completes several runs of Metropolis-Hastings simulations (as many as determined in the option mh nblocks. with the first. ESTIMATING DSGE MODELS . For the results to be sensible.58 CHAPTER 5. In our example above. try starting with a uniform and relatively wide prior and see where the data leads the posterior distribution. And second. results between the various chains should be close. More specifically. and the optimizer did not get stuck in an odd area of the parameter subspace. First. two things should happen. you can also . Dynare plots a green line which represents the posterior mode.figure 6 in our example . with results. we obtain convergence and relative stability in all measures of the parameter moments. these should be relatively constant (although there will always be some variation) and they should converge. TIP! If the plotted moments are highly unstable or do not converge. If you have trouble coming up with a new prior.displays the most interesting set of results. In each case. the red and blue lines on the charts represent specific measures of the parameter vectors both within and between chains. In addition. being a measure of the variance and “m3” based on third moments. or spot a problem. and the vertical axis the measure of the parameter moments. In fact.

The horizontal axis. The last figure returns the smoothed estimated shocks in a useful illustration to eye-ball the plausibility of the size and frequency of the shocks. the green mode (calculated from the numerical optimization of the posterior kernel) should not be too far away from the mode of the posterior distribution. Second. Third. TIP! These graphs are of course especially relevant and present key results. the posterior distributions should be close to normal. or at least not display a shape that is clearly non-normal.5. . First. but they can also serve as tools to detect problems or build additional confidence in your results. the prior and the posterior distributions should not be excessively different.9. in this case. One thing to check is the fact that shocks should be centered around zero. it is advisable to undertake a greater number of Metropolis-Hastings simulations. INTERPRETING OUTPUT 59 discuss the probability that your parameter is larger or smaller than a certain value. If not. That is indeed the case for our example. represents the number of periods in the sample.

.

1 Introducing the example The example is drawn from Schorfheide (2000). while emphasizing techniques to deal with non-stationary observations and stochastic trends in dynamics. 6. Note that the original paper by Schorfheide mainly focusses on estimation methodologies. The chapter begins by presenting a more complex example than the one used for illustration purposes in chapter 5. its basic intuitions and equations. with a special interest in model comparison. The example shows how to use Dynare in a more realistic setting. The chapter then follows with sections on comparing models to one another.Chapter 6 Estimating DSGE models advanced topics This chapter focusses on advanced topics and features of Dynare in the area of model estimation. The goal is to show how Dynare would be used in the more “realistic” setting of reproducing a recent academic paper. difficulties and solutions. and ends with a table summarizing where output series are stored and how these can be retrieved. So we thought you would enjoy (and continue learning from!) a more realistic example which reproduces the work in a recent .1 Alternative and non-stationary example The example provided in chapter 5 is really only useful for illustration purposes. We will then see in subsequent sections how to estimate it using Dynare.academic paper. and then to BVARs. This first section introduces the model. That paper should serve as a helpful reference if anything is 61 . 6.1.and highly regarded . while the mathematics and economic intuitions of the model it evaluates are drawn from Nason and Cogley (1994).

ESTIMATING DSGE MODELS .1 sketches the main dynamics of the model. Ct . Dt . there are several markets and actors to keep track of. left unclear in the description below. In essence. In the model. in order to earn RH. So to clarify things. Households maximize their utility function which depends on consumption. and a monetary authority which plays a minor role. Mt+1 and how much to deposit at the bank.1. figure 6. Ht .62 CHAPTER 6. while deciding how much money to hold next period in cash.1: Continuous lines show the circulation of nominal funds. The goal of the paper is to estimate the model using Bayesian techniques. Households . and hours worked.ADVANCED TOPICS Figure 6. You may want to refer back to the figure as you read through the following sections. while dashed lines show the flow of real variables. The economy is made up of three central agents and one secondary agent: households. the model studied by Schorfheide (2000) is one of cash in advance (CIA).t − 1 interest. while observing only output and inflation. firms and banks (representing the financial sector).

It uses these funds to disburse loans to firms. Finally. Mt+1 − Mt ). where It = Kt+1 − (1 − δ)Kt . next period’s capital stock. Ft ≤ Lt + Pt Kt (At Nt )1−α − Kt+1 + (1 − δ)Kt − Wt Nt − Lt RF. Pt Ct ≤ Mt − Dt + Wt Ht Mt+1 = (Mt − Dt + Wt Ht − Pt Ct ) + RH. as well as dividends from firms. To close the model.t Dt −Lt +Xt . and loans. Its problem is summarized by max {Ft .t. Bt are simply equal to Dt +RF. wages. and where δ is the rate of depreciation.t.t Dt + Ft + Bt where the second equation spells out the cash in advance constraint including wage revenues. Lt . receive cash deposits from households and a cash injection. by trading off investment for dividends to consumers. which in both cases are made up of net cash inflows defined below. interests. Finally. bank dividends. Of course.1. Ft .Nt . banks are constrained in their loans by a credit market equilibrium condition Lt ≤ Xt + Dt . Xt from the central bank (which equals the net change in nominal money balances. The third equation simply specifies that bank loans are used to pay for wage costs.t due to the equal risk profiles of the loans.Lt } E0 ∞ Ft t+1 t=0 β Ct+1 Pt+1 α s.t Lt −RH. and from banks.6. Bt .Dt } 63 E0 ∞ t t=0 β [(1 − φ) ln Ct + φ ln(1 − Ht )] 0 ≤ Dt s. on which they make a net return of RF.t Wt Nt ≤ Lt α where the second equation makes use of the production function Yt = Kt (At Nt )1−α and the real aggregate accounting constraint (goods market equilibrium) Ct + It = Yt .Mt+1 . .t = RF. Kt+1 .Kt+1 . we add the usual labor and money market equilibrium equations. the third the inability to borrow from the bank and the fourth the intertemporal budget constraint emphasizing that households accumulate the money that remains after bank deposits and purchases on goods are deducted from total inflows made up of the money they receive from last period’s cash balances. Ht = Nt and Pt Ct = Mt + Xt . ALTERNATIVE AND NON-STATIONARY EXAMPLE therefore solve the problem max {Ct . firms maximize the net present value of future dividends (discounted by the marginal utility of consumption. Note that it is the firms that engage in investment in this model. Nt . on their end. labor demand. Banks.Ht . as well as the condition that RH. since they are owned by households) by choosing dividends.t − 1.

we obtain the following set of first order and equilibrium conditions.t .t . A. if the logs of a variable are specified to follow a unit root process. affecting technology and one nominal. Note that theses expressions for trends are not written in the most straightforward manner nor very consistently.t 2 ∼ N (0. These important equations are ln At = γ + ln At−1 + A. but encourage you to browse Nason and Cogley (1994) for additional details. σM ) where mt ≡ MT +1 /Mt is the growth rate of the money stock. Otherwise said. As usual. for details. and the second an autoregressive stationary process in the growth rate of money.1 When the above functions are maximized. we could have written the AR(2) process in state space form and realized that the system has an eigenvalue of one. ESTIMATING DSGE MODELS .64 CHAPTER 6. We nonetheless give a brief intuitive explanation of each 1 Alternatively. but an AR(2) with a unit root in the log of the level of money.ADVANCED TOPICS More importantly. The model allows for two sources of perturbations. affecting the money stock. to save space. chapter 15. one is a root of the second order autoregressive lag polynomial. we add a stochastic elements to the model. . The first equation is therefore a unit root with drift in the log of technology.t 2 ∼ N (0. We will not dwell on the derivations here. M. the rate of growth of the series is a stationary stochastic process. one real. see Hamilton (1994). This can be seen from the definition of mt which can be rewritten as ln Mt+1 = ln Mt + ln mt . But we reproduced them never-the-less to make it easier to compare this example to the original paper. σA ) and ln mt = (1 − ρ) ln m∗ + ρ ln mt−1 + M.

and the marginal rate of substitution between consumption and leisure.t+1 ) −1 A. ALTERNATIVE AND NON-STATIONARY EXAMPLE equation. also affecting labor demand. 2. In order.1. as firms use borrowed funds to pay for labor input.t ) Yt /Yt−1 A. The firms’ borrowing constraint. these equations correspond to: 1. 3.t Pt /Pt−1 = (Pt /Pt−1 )(mt−1 /eγ+ ) where. hats over variables no longer mean deviations from steady state.6.t+1 ) 65 α−1 1−α Pt+1 αKt Nt+1 + (1 − δ) / ct+2 Pt+2 mt+1 Wt = Lt /Nt φ Ct Pt / (1 − Nt ) 1−φ −1 = Lt /Nt A. labor demand. but instead represent variables that have been made stationary.t ) α Kt−1 N 1−α + (1 − δ)e−(γ+ Kt−1 α Yt = Kt−1 N 1−α e−α(γ+ A.3 below. 4.t M. The Euler equation in the goods market. we pause a moment to give some intuition for the above equations. representing the tradeoff to the economy of moving consumption goods across time. The intertemporal labor market optimality condition. importantly. . The equilibrium interest rate in which the marginal revenue product of labor equals the cost of borrowing to pay for that additional unit of labor. We come back to this important topic in details in section 6.t ) ln(mt ) = (1 − ρ) ln(m ) + ρ ln(mt−1 ) + At ≡ dAt = exp(γ + At−1 Yt /Yt−1 = eγ+ A.t ) Ct + Kt = e−α(γ+ Pt C = mt mt − 1 + Dt = Lt A.1. linking labor supply.t+1 ) Rt = (1 − α)Pt e−α(γ+ Ct P t α Kt−1 Nt−α /Wt α Kt−1 Nt1−α = β (1 − α) Pt e−α(γ+ ×Et Lt mt Ct+1 Pt+1 A. For now.t A. The system comes down to Et − Pt / Ct+1 Pt+1 mt = βe−α(γ+ A.

2 Declaring variables and parameters This block of the . The Euler equation in the credit market. real variables grow with At (except for labor. Nt .1. 7.mod file follows the usual conventions and would look like: var m P c e W R k d n l Y obs P obs y dA. It can be shown that when shocks are null. where the choice of upper and lower case letters is not significant.3 The origin of non-stationarity The problem of non-stationarity comes from having stochastic trends in technology and money. The stochastic process for technology. 9. 6. 11. del. 6. 6. The exogenous variables are as expected and concern the shocks to the evolution of technology and money balances.1. mst. varexo e a e m. The credit market equilibrium condition. The stochastic process for money growth. the first set of endogenous variables. 12. psi.66 CHAPTER 6. which is stationary as there is no population growth). The non-stationarity comes out clearly when attempting to solve the model for a steady state and realizing it does not have one. The aggregate resource constraint. 10. Detrending therefore involves . which ensures that giving up one unit of consumption today for additional savings equals the net present value of future consumption. The money market equilibrium condition equating nominal consumption demand to money demand to money supply to current nominal balances plus money injection. nominal variables grow with Mt and prices with Mt /At . gam. rho. up to l. bet. 8. The production function. parameters alp. ESTIMATING DSGE MODELS . are as specified in the model setup above.ADVANCED TOPICS 5. The relationship between observable variables and stationary variables. more details on these last two equations appear in the following section. and where the last five variables are defined and explained in more details in the section below on declaring the model in Dynare.

1. and leave the transformations of the remaining equations as an exercise. . Pt = Pt ·At /Mt . in the output equation.5 in chapter 3). For nominal ˆ ˆ t = Qt /Mt . lt .mod file after transforming them in exactly the same way as the one above.6.4 Stationarizing variables Let’s illustrate this transformation on output. ALTERNATIVE AND NON-STATIONARY EXAMPLE 67 the following operations (where hats over variables represent stationary variables): for real variables.mod file of Dynare into which we enter: y=k(-1)^alp*n^(1-alp)*exp(-alp*(gam+e a)) The other equations are entered into the . where qt = [yt .t )) where we go from the second to the third line by taking the exponential of both sides of the equation of motion of technology. The resulting equation made up of stationary variables is Yt At = Kt−1 At−1 α 1−α At Nt1−α A−1 Aα t−1 t α Yt = Kt−1 Nt1−α At At−1 −α α = Kt−1 Nt1−α exp(−α(γ + A. we should actually work with Kt−1 = Kt−1 /At−1 . We define Yt to equal Yt /At and Kt as Kt /At . ˆ variables. A final transformation to consider. kt+1 ]. Q 6. Wt ]. NOTE! Recall from section 3.1. where Qt = [dt . that in Dynare variables take the time subscript of the period in which they are decided (in the case of the capital stock. qt = qt /At . ct . The above is the equation we retain for the . today’s capital stock is a result of yesterday’s decision). that turns out to be useful since we often deal with the growth rate of technology. We stationarize output by dividing its real variables (except for labor) by At . Thus. is to define dA = exp(gam+e a) by simply taking the exponential of both sides of the stochastic process of technology defined in the model setup above. if you wish (Nason and Cogley (1994) includes more details on the transformations of each equation). it . And for prices.

We could simply write Yobs = Yt At .1. But. We would then have to relate this observable. We could simply stationarize them by working with rates of growth (which we know are constant). where. the observable variable would become Yt /Yt−1 . In the case of output. we must make a decision as to our non-stationary observations. To the . The goal is to add a line to the model block of our . we therefore follow this path in the remainder of the example. the y of the . Yt . This complicates things somewhat. 1/(c*P)-bet*P*(1-alp)*exp(-alp*(gam+e a))*k(-1)^alp*n^(1-alp)/ (m*l*c(+1)*P(+1)) = 0. The result is not very different. though.5 Linking stationary variables to the data And finally.mod file model. we add to the model block of the . but illustrates several features of Dynare worth highlighting.6 The resulting model block of the . c+k = exp(-alp*(gam+e a))*k(-1)^alp*n^(1-alp)+(1-del) . call them Yobs . W = l/n. -P/(c(+1)*P(+1)*m)+bet*P(+1)*(alp*exp(-alp*(gam+log(e(+1))))*k^(alp-1) *n(+1)^(1-alp)+(1-del)*exp(-(gam+log(e(+1)))))/(c(+2)*P(+2)*m(+1))=0. to our (stationary) model’s variables Yt by using the definition that Yt ≡ Yt /At .mod file are the stationary Yt .1. Thus. ESTIMATING DSGE MODELS . we we-write the above relationship in ratios. dA = exp(gam+e a).68 CHAPTER 6. -(psi/(1-psi))*(c*P/(1-n))+l/n = 0. but just a dAt . R = P*(1-alp)*exp(-alp*(gam+e a))*k(-1)^alp*n^(-alp)/W. we could also work with non-stationary data in levels. 6.mod file. call it gy obs. But since we don’t have an At variable. from what we just saw above. log(m) = (1-rho)*log(mst) + rho*log(m(-1))+e m. we therefore add: Y obs/Y obs(-1) = dA*y/y(-1). except that we use the relationship Pobs = Pt Mt /At as noted earlier. our other observable variable.mod file: gy obs = dA*y/y(-1). The details of the correct transformations for prices are left as an exercise and can be checked against the results below.ADVANCED TOPICS 6. to our stationary output.mod file that relates the non stationary observables. We of course do the same for prices.

P obs/P obs(-1) = (p/p(-1))*m(-1)/dA. e = exp(e a). after all.7 Declaring observable variables We begin by declaring which of our model’s variables are observables. when dealing with non-stationary observations. are the same as those listed in chapter 3. as the last two equations don’t have a steady state. Thus. to specify that our observable variables are indeed P obs and Y obs as noted in the section above. you must loglinearize your model (and not just linearize it). 69 where. This can be seen explicitly by taking the difference of logs of output and prices to compute growth rates. non-stationary. concerning the observables which are. The problem that arises. If we first take logs. notice that what we have done is in fact relegated the non-stationarity of the model to just the last two equations. though.1. is that we cannot linearize the above system in levels. this is a point to which we will return later. P*c = m.1.1.mod file we write varobs P obs Y obs. If this is not the case. In the case of output. though.8 Declaring trends in observable variables Recall that we decided to work with the non-stationary observable variables in levels. m-1+d = l. 6. of course. NOTE! Recall from earlier that the number of observed variables must be smaller or equal to the number of shocks such that the model be estimated. ALTERNATIVE AND NON-STATIONARY EXAMPLE *exp(-(gam+e a))*k(-1). they become linear and it doesn’t matter anymore where we calculate their derivative when taking a Taylor expansion of all the equations in the system. In our . 6. y = k(-1)^alp*n^(1-alp)*exp(-alp*(gam+e a)). such as ending lines with semicolons and indicating the timing of variables in parentheses. Y obs/Y obs(-1) = dA*y/y(-1). the input conventions. Both output and prices exhibit stochastic trends. we make use of the usual (by now!) . end.6. you should add measurement shocks to your model where you deem most appropriate. TIP! In the above model block.

since P obs and Y obs inherit the unit root characteristics of their driving variables. Taking logs of both sides and subtracting the same equation scrolled back one period. so that the drift terms in the above equation are ln m∗ − γ.ADVANCED TOPICS relationship Yt = Yt · At . any trends.1. whereas we know ∆ ln Yt is stationary in steady state. Unit roots are only related to stochastic trends. we write in the . ln mt = ln m∗ . end. we apply the same manipulations to show that: ∆ ln Pt = ∆ ln Pt + ln mt−1 − γ − A. To give these instructions to Dynare.t − γ − A.2 In Dynare. NOTE! You don’t need to declare unit roots for any non-stationary model. the command observation trends specifies linear trends as a function of model parameters for the observed variables in the model.t Note from the original equation of motion of ln mt that in steady state.t emphasizing clearly the drift term γ.mod unit root vars P obs Y obs. Y obs (gam). In general. Note that for stationary variables. whether deterministic or stochastic (the drift term) must be declared up front. technology and money. In the case of prices. except for ln m∗ and γ. 2 . P obs (log(mst)-gam). 6. the unconditional covariance matrix of these variables is used for initialization. In the case of our example. we must tell Dynare to use a diffuse prior (infinite variance) for their initialization in the Kalman filter.70 CHAPTER 6. The algorithm to compute a true diffuse prior is taken from Durbin and Koopman (2001). we find: ∆ ln Yt = ∆ ln Yt + γ + A. we therefore write (in a somewhat cumbersome manner) observation trends. You don’t need to use a diffuse This can also be see from substituting for ln mt−1 in the above equation with the equation of motion of ln mt to yield: ∆ ln Pt = ∆ ln Pt + ln m∗ + ρ(ln mt−2 − ln m∗ ) + M.t where all terms on the right hand side are constant.9 Declaring unit roots in observable variables And finally. ESTIMATING DSGE MODELS .

65. 0. beta pdf.002. 1.5.m. First.0085. The file is called fs2000ns steadystate.993. nothing has changed with respect to the equivalent file of chapter 5.356. P obs and Y obs which take the value of 1.007.1. 6. 0. psi. intuitive. ALTERNATIVE AND NON-STATIONARY EXAMPLE 71 initial condition in the case of a deterministic trend. 0. In this case. covering the initval.10 Specifying the steady state Declaring the steady state is just as explained in details and according to the same syntax explained in chapter 3. gam. 0. normal pdf. stderr e m. notice the declaration of parameters at the beginning. del. 0. 0.223. inf. 0. inf. 0. 6. alp. but tedious.6. end. The check=0 limits steady state values to real numbers. This functionality may be updated in later versions of Dynare. normal pdf. the output of the function is the endogenous variables at steady state. 0..11 Declaring priors We expand our .1.05. So far. steady and check commands. beta pdf. rho. section 5. defining each variable in terms of parameters or variables solved in the lines above. 0.1. The only novelty is the declaration of the non-stationary variables. beta pdf.1. we also discussed the usefulness of providing an external Matlab file to solve for the steady state.01. 0.005. bet.12 Launching the estimation We add the following commands to ask Dynare to run a basic estimation of our model: .0002. the ys vector.003. This is Dynare convention and must be the case for all your non-stationary variables. beta pdf. 6. In chapter 5..129. inv gamma pdf. inv gamma pdf. mst. Second. stderr e a. you can find the corresponding steady state file in the models folder under UserGuide. Third. 0. 0. There are some things to notice.035449.008862.mod file with the following information: estimated params. 0. since the variance is finite. note that the file is really only a sequential set of equalities. 0. beta pdf.02.

45. P obs/P obs(-1) = (p/p(-1))*m(-1)/dA. For convenience. mode compute=6.13 The complete . R = P*(1-alp)*exp(-alp*(gam+e a))*k(-1)^alp*n^(-alp)/W. it’s now time to get a picture of what the complete file looks like.mh jscale=0. varobs P obs Y obs. A simple linearization would fail as these variables do not have a steady state.72 CHAPTER 6.mod. var m P c e W R k d n l Y obs P obs y dA.loglinear. NOTE! As mentioned earlier. P*c = m. we need to instruct Dynare to log-linearize our model. mst. parameters alp. c+k = exp(-alp*(gam+e a))*k(-1)^alp*n^(1-alp)+(1-del) exp(-(gam+e a))*k(-1). varexo e a e m.nobs=192. y = k(-1)^alp*n^(1-alp)*exp(-alp*(gam+e a)).65). taking the log of the equations involving non-stationary variables does the job of linearizing them. 6. since it contains non-linear equations in non-stationary variables. model. Fortunately.mh nblocks=2. dA = exp(gam+e a). ESTIMATING DSGE MODELS .mod file We have seen each part of the . m-1+d = l. 1/(c*P)-bet*P*(1-alp)*exp(-alp*(gam+e a))*k(-1)^alp*n^(1-alp)/ (m*l*c(+1)*P(+1)) = 0. The file is called fs2000ns. -P/(c(+1)*P(+1)*m)+bet*P(+1)*(alp*exp(-alp*(gam+log(e(+1)))) *k^(alp-1)*n(+1)^(1-alp)+(1-del) *exp(-(gam+log(e(+1)))))/(c(+2)*P(+2)*m(+1))=0.ADVANCED TOPICS estimation(datafile=fsdat. gam. log(m) = (1-rho)*log(mst) + rho*log(m(-1))+e m. -(psi/(1-psi))*(c*P/(1-n))+l/n = 0. rho. W = l/n. e = exp(e a).mh replic=2000. Y obs/Y obs(-1) = dA*y/y(-1).mod separately. the file also appears in the models folder under UserGuide in your Dynare installation.mh drop=0.1. . bet. del. psi. end.

ALTERNATIVE AND NON-STATIONARY EXAMPLE observation trends.mh nblocks=2. inv gamma pdf.mh drop=0. beta pdf. end. alp. estimation(datafile=fsdat.223. 0. e = 1.02. 0.loglinear. stderr e a. psi. del. inf. estimated params. R = 1. initval. k = 6. m = mst.mh jscale=0. l = 0.035449. steady. P = 2. 0. 0.6.6. inf.65.002. end. 73 // the above is really only useful if you want to do a stoch simul // of your model.356. 0.0085.01.0002. gam.1. end.02. beta pdf. mst. y = 0. normal pdf.008862. 0.05.005. c = 0. stderr e m. P obs (log(mst)-gam). 0. beta pdf. W = 4. 0. normal pdf.45. 0.mh replic=2000. d = 0. mode compute=6.86.129.25.nobs=192.993. 0.19. beta pdf. 0. 1. unit root vars P obs Y obs. since the estimation will use the Matlab // steady state file also provided and discussed above. dA = exp(gam).45. 0. 0.003. bet. rho. 0.65). inv gamma pdf. n = 0. 0. Y obs (gam).85. . beta pdf.007.

there are five basic steps to translate a model into Dynare for successful estimation. ESTIMATING DSGE MODELS . It may therefore be useful. to gain a “bird’s eyeview” on what we have just accomplished. 6.1. 6.1. This is done in figure 6.14.2: At a high level.ADVANCED TOPICS Figure 6.74 CHAPTER 6.2 Comparing models based on their posterior distributions ** TBD . and summarize the most important steps at a high level.14 Summing it up The explanations given above of each step necessary to translate the Schorfheide (2000) example into language that Dynare can understand and process was quite lengthy and involved a slew of new commands and information.

WHERE IS YOUR OUTPUT STORED? 75 6.6.3. The Reference Manual overviews the complete set of potential output files and describes where you can find each one. . depending on the instructions you give Dynare.3 Where is your output stored? The output from estimation can be extremely varied.

.

note that in this chapter we will focus on stochastic models . or will spur your curiosity to have a look at more detailed material. and when expectations are taken. you may want to go through Michel Juillard’s presentation on solving DSGE models to a first and second order (available on Michel Juillard’s website). Finally. or under its hood. linearizing a system of equations to the first order raises the issue of certainty equivalence.1. Thus.Chapter 7 Solving DSGE models Behind the scenes of Dynare 7. as explained in section 3. This is because only the first moments of the shocks enter the linearized equations. to get an idea of the methodologies and algorithms used in its computations.1 Introduction The aim of this chapter is to peer behind the scenes of Dynare. For more details on the Newton-Raphson algorithm used in Dynare to solve deterministic models.1 of chapter 3. see Juillard (1996). 77 . What you will find below will either comfort you in realizing that Dynare does what you expected of it and what you would have also done if you had had to code it all yourself (with a little extra time on your hands!). Going into details would be beyond the scope of this User Guide which will instead remain at a high level.2 What is the advantage of a second order approximation? As noted in chapter 3 and as will become clear in the section below. or read Collard and Juillard (2001a) or Schmitt-Grohe and Uribe (2004) which gives a good overview of the most recent solution techniques based on perturbation methods. 7.which is where the major complication lies. they disappear. If so.

a DSGE model is a collection of first order and equilibrium conditions that take the general form: Et {f (yt+1 . The summary below is taken mainly from Michel Juillard’s presentation “Computing first order approximations of DSGE models with Dynare”. Yet more clearly. This may be an acceptable simplification to make. For instance. SOLVING DSGE MODELS . linearizing to the second order enables you to take risk (or the variance of shocks) into consideration . you also need second order approximations of the policy function. that satisfy the original system. we shall briefly overview the perturbation methods employed by Dynare to solve DSGE models to a first order approximation.BEHIND THE SCENES OF 78 DYNARE unconditional expectations of the endogenous variables are equal to their nonstochastic steady state values. it may instead be quite misleading. ut ) .a highly desirable modeling feature. when using second order welfare functions to compare policies. Sticking to the above notation. which you should read if interested in particular details. yt−1 . 7. It is therefore very convenient that Dynare allows you to choose between a first or second order linearization of your model in the option of the stoch simul command.CHAPTER 7. although at a higher level of complexity. This is what we call the policy function. But depending on the context. ut )} = 0 E(ut ) = 0 E(ut ut ) = Σu and where: y : vector of endogenous variables of any dimension u : vector of exogenous stochastic shocks of any dimension The solution to this system is a set of equations relating variables in the current period to the past state of the system and current shocks. To summarize. The second order follows very much the same approach. we can write this function as: yt = g(yt−1 . especially regarding second order approximations (available on Michel Juillard’s website). yt . in the case of asset pricing models.3 How does dynare solve stochastic DSGE models? In this section.

ut+1 ) = Et f (¯. y . ut+1 )] = 0 We then venture to linearize this model around a steady state defined as: f (¯. ut ). ut . and current and future shocks: Et [F (yt−1 . g(yt−1 . 0) = 0 y ¯ ¯ having the property that: y = g(¯. ut ) which enables us to rewrite our system in terms of past variables.3. y . u = ut . they drop out when taking expectations of the linearized equations. 0) ¯ y The first order Taylor expansion around y yields: ¯ Et F (1) (yt−1 . yt−1 . This is technically why certainty equivalence holds . ut . ut+1 ) = f (¯. ut+1 ) = f (g(g(yt−1 . y . fy+ = ˆ ¯ fu = ∂f ∂ut . ut+1 ). ut+1 ) 79 We can then define a new function F . Taking expectations (we’re almost there!): Et F (1) (yt−1 . gy = ∂g ∂yt−1 . fy0 = ∂f ∂yt . y ) + fy+ (gy (gy y + gu u)) y ¯ ¯ ˆ +fy0 (gy y + gu u) + fy− y + fu u ˆ ˆ = fy+ gy gy + fy0 gy + fy− y + fy+ gy gu + fy0 gu + fu u ˆ = 0 As you can see. ut . since future shocks only enter with their first moments (which are zero in expectations). gu = ∂g ∂ut . y .7. ut ). fy− = ∂f ∂yt−1 . u = ut+1 . y ) + fy+ gy (gy y + gu u) + gu u y ¯ ¯ ˆ +fy0 (gy y + gu u) + fy− y + fu u ˆ ˆ =0 with y = yt−1 − y . ut ). such that: F (yt−1 . HOW DOES DYNARE SOLVE STOCHASTIC DSGE MODELS? Then. ut+1 ) = g(g(yt−1 . ∂f ∂yt+1 . ut . it is straightforward to re-write yt+1 as yt+1 = g(yt .

The second thing to note is that we have two unknown variables in the above equation: gy and gu each of which will help us recover the policy function g. Note that in the case of a second order approximation of a DSGE model. which we can solve with a series of algebraic trics that are not all immediately apparent (but detailed in Michel Juillard’s presentation). SOLVING DSGE MODELS .BEHIND THE SCENES OF 80 DYNARE in a system linearized to its first order. Incidentally. for instance. The first. If we were interested in impulse response functions. . each parenthesis must ˆ be null and we can solve each at a time. recovering gu is then straightforward from the second parenthesis.CHAPTER 7. the variance of future shocks remains after taking expectations of the linearized equations and therefore affects the level of the resulting policy function. yields a quadratic equation in gy . Since the above equation holds for any y and any u. we would simply iterate the policy function starting from an initial value given by the steady state. we have solved for the (approximate) policy (or decision) function and have succeeded in solving our DSGE model. The second order solution uses the same “perturbation methods” as above (the notion of starting from a function you can solve . one of the conditions that comes out of the solution of this equation is the Blanchard-Kahn condition: there must be as many roots larger than one in modulus as there are forward-looking variables in the model. Finally. notice that a first order linearization of the function g yields: yt = y + gy y + gu u ¯ ˆ And now that we have gy and gu . But the general approach is perfectly isomorphic. Having recovered gy .like a steady state and iterating forward). yet applies more complex algebraic techniques to recover the various partial derivatives of the policy function.

using the Kalman filter to find the likelihood function. Lubik and Schorfheide (2005) which applies Bayesian estimation methods to an open macro model focussing on issues of misspecification and identification. It begins by motivating Bayesian estimation by suggesting some arguments in favor of it as opposed to other forms of model estimation. FernandezVillaverde and Rubio-Ramirez (2004) which reviews the econometric properties of Bayesian estimators and compare estimation results with maximum likelihood and BVAR methodologies. 81 . It then attempts to shed some light on what goes on in Dynare’s machinery when it estimates DSGE models. based on posterior distributions. this section surveys the methodologies adopted for Bayesian estimation. and finally Rabanal and Rubio-Ramirez (2005) which compares the fit. including defining what are prior and posterior distributions. and comparing models based on posterior distributions. 8. some of these include: Schorfheide (2000) which uses Bayesian methods to compare the fit of two competing DSGE models of consumption. Smets and Wouters (2003) which applies Bayesian estimation techniques to a model of the Eurozone. Recent papers have attracted significant attention. estimating the posterior function thanks to the Metropolis-Hastings algorithm. of four competing specifications of New Keynesian monetary models with nominal rigidities. Lubik and Schorfheide (2003) which investigates whether central banks in small open economies respond to exchange rate movements. Ireland (2004) which emphasizes instead maximum likelihood estimation.Chapter 8 Estimating DSGE models Behind the scenes of Dynare This chapter focuses on the theory of Bayesian estimation.1 Advantages of Bayesian estimation Bayesian estimation is becoming increasingly popular in the field of macroeconomics. To do so.

as opposed to GMM estimation which is based on particular equilibrium relationships such as the Euler equation in consumption. estimation in the Bayesian case is based on the likelihood generated by the DSGE system. Bayesian estimation naturally leads to the comparison of models based on fit. First. solved DSGE model. when estimating a model. as other topics mentioned above. This procedure. Sixth. Bayesian estimation explicitly addresses model misspecification by including shocks. Likewise. Bayesian estimation fits the complete. the inclusion of priors also helps identifying parameters. in the structural equations. 8. Indeed. due to the stylized and often misspecified nature of DSGE models. Indeed. St´phane Adjemian.BEHIND THE SCENES 82 OF DYNARE There are a multitude of advantages of using Bayesian methods to estimate a model. estimating it using Bayesian techniques could be a disadvantage.2 The basic mechanics of Bayesian estimation This and the following subsections are based in great part on work by. rather than the more indirect discrepancy between the implied DSGE and VAR impulse response functions. More technically. Unfortunately. and discussions with. Fourth. But the weighting of the likelihood with prior densities often leads to adding just enough curvature in the posterior distribution to facilitate numerical maximization. is discussed more technically in the subsection below. a member of the Dynare development e . Of course. Bayesian techniques allow the consideration of priors which work as weights in the estimation process so that the posterior distribution avoids peaking at strange points where the likelihood peaks. the likelihood often peaks in regions of the parameter space that are contradictory with common observations. It can be summarized by different values of structural parameters leading to the same joint distribution for observables. the posterior distribution corresponding to competing models can easily be used to determine which model best fits the data. the problem arises when the posterior distribution is flat over a subspace of parameter values. leading to the “dilemma of absurd parameter estimates”. ESTIMATING DSGE MODELS . Second. but five of these stand out as particularly important and general enough to mention here. Third.CHAPTER 8. the problem of identification often arises. which can be interpreted as observation errors. if your model is entirely mis-specified.

Second. these two building blocks .mod files touching on Bayesian estimation. generalized beta. At its most basic level. which includes a clear and quite complete introduction to Bayesian estimation. Finally. introduction to the basic mechanics of Bayesian estimation. although summarized in presentation format. or uniform function.2. a likelihood p(YT |θ). the posterior density. θ A represents the parameters of model A. A) = p(y0 |θ A . beta. Also.are tied together by Bayes’ rule. Generally speaking. THE BASIC MECHANICS OF BAYESIAN ESTIMATION 83 team. the appendix of Schorfheide (2000) contains details as to the exact methodology and possible difficulties encountered in Bayesian estimation. In the end. Finally. A) where YT are the observations until period T .priors and likelihood functions . Some of this work. given the model and its parameters: L(θ A |YT . The tradition of calibrating models is inherited through the specification of priors. p(•) stands for a probability density function (pdf) such as a normal. And the maximum likelihood approach enters through the estimation process based on confronting the model with data. Bayesian estimation is a bridge between calibration and maximum likelihood. Using the Bayes theorem . gamma. First. priors are described by a density function of the form p(θ A |A) where A stands for a specific model. More technically. the likelihood function describes the density of the observed data. we have a prior density p(θ) on the one hand. which provides a very clear. θ A . although somewhat outdated. and on the other. A) t=1 p(yt |Yt−1 . is available in the “Events” page of the Dynare website. inverse gamma. A) We now take a step back. shifted gamma. Together. from example files to lecture notes to related papers. priors can be seen as weights on the likelihood function in order to give more importance to certain areas of the parameter subspace. You may also want to take a glance at Hamilton (1994). and where in our case the likelihood is recursive and can be written as: T p(YT |θ A . we are interested in p(θ|YT ). Other helpful material includes An and Schorfheide (2006). the websites of Frank Schorfheide and Jesus Fernandez-Villaverde contain a wide variety of very helpful material.8. A) ≡ p(YT |θ A . illustrated by the application of a simple DSGE model. chapter 12. Let’s see how. remember to also check the open online examples of the Dynare website for examples of .

. A)p(θ A |A) ≡ K(θ A |YT . A) = p(YT |θ A . 8. we have p(θ. 1) is gaussian white noise. though.CHAPTER 8. ESTIMATING DSGE MODELS . we can combine the prior density and the likelihood function discussed above to get the posterior density: p(θ A |YT . YT ) = p(YT |θ) × p(θ) p(θ) By using these identities.an example Suppose a data generating process yt = µ + εt for t = 1. T . the likelihood is given by p(YT |µ) = (2π)− 2 e− 2 We know from the above that µM L.. YT ) p(θ|YT ) = p(YT ) We also know that p(YT |θ) = p(θ. the posterior kernel (or un-normalized posterior density. YT ) ⇔ p(θ. Then. chapter 12. YT |A)dθ A Finally. The example is drawn from Zellner (1971). A) This is the fundamental equation that will allow us to rebuild all posterior moments of interest. corresponds to the numerator of the posterior density: p(θ A |YT .. The trick will be to estimate the likelihood function with the help of the Kalman filter and then simulate the posterior kernel using a sampling-like or Monte Carlo method such as the Metropolis-Hastings. Before moving on. A)p(θ A |A) p(YT |A) where p(YT |A) is the marginal density of the data conditional on the model: p(YT |A) = ΘA p(θ A . although other similar examples can be found in Hamilton (1994). . where εt ∼ N (0. Generally.T ] = 1 T. A) ∝ p(YT |θ A .BEHIND THE SCENES 84 OF DYNARE twice we obtain this density of parameters knowing the data.1 Bayesian estimation: somewhere between calibration and maximum likelihood estimation . .2. the subsection below gives a simple example based on the above reasoning of what we mean when we say that Bayesian estimation is “somewhere in between calibration and maximum likelihood estimation”.T = 1 T T 1 T 2 t=1 (yt −µ) T t=1 yt ≡ y and that V[µM L. given that the marginal density above is a constant or equal for any parameter). These topics are covered in more details below.

so we just estimate the model) then E[µ] → µM L. we have no prior information. the posterior density is defined. ut )} = 0. with 1 −2 + σµ V[µ] = and E[µ] = 1 −1 T 1 −1 −2 µM L. p (µ|YT ) ∝ e − (µ−E[µ])2 V[µ] . yt−1 . In more appropriate terms for what follows. Other variables are described below. y is the vector of steady ˆ ¯ state values and θ the vector of deep (or structural) parameters to be estimated. In particular. 8. But if σµ → 0 (ie. The second equation is the familiar decision rule mentioned above.3 8. can be written in the form Et {f (yt+1 . But the equation expresses a relationship among true endogenous variables that ∗ are not directly observed. the prior mean.3. ut ). we’re sure of ourselves and we calibrate the parameter of interest. taking as a solution equations of the type yt = g(yt−1 . if σµ → ∞ (ie. which is really a collection of first order and equilibrium conditions. the 2 maximum likelihood estimator. Then. Only yt is observable.8. DSGE MODELS AND BAYESIAN ESTIMATION 85 In addition. Most of the time. yt . and it is related to the true . which we call the decision rule.T .1 DSGE models and Bayesian estimation Rewriting the solution to the DSGE model Recall from chapter 7 that any DSGE model. we can tell that the posterior mean is a convex combination of 2 the prior mean and the ML estimate. let our prior be a gaussian distribution with expectation µ0 2 and variance σµ . we can rewrite the solution to a DSGE model as a system in the following manner: ∗ yt = M y (θ) + M yt + N (θ)xt + ηt ¯ ˆ yt = gy (θ)ˆt−1 + gu (θ)ut ˆ y E(ηt ηt ) = V (θ) E(ut ut ) = Q(θ) where yt are variables in deviations from steady state. up to a constant. we’re somewhere in the middle of these two extremes.3. by: p (µ|YT ) ∝ 1 1 − 2 (2πσµ )− 2 e 2 (µ−µ0 )2 2 σµ × (2π)− 2 e− 2 T 1 T 2 t=1 (yt −µ) Or equivalently. thus leaving no room for estimation) then E[µ] → µ0 .T + σµ µ0 T −2 1 −1 + σµ T From this.

BEHIND THE SCENES 86 OF DYNARE variables with an error ηt . ESTIMATING DSGE MODELS . . The log-likelihood above gets us one step closer to our goal of finding the posterior distribution of our parameters. Furthermore.3. though. From the Kalman filter recursion. 8. which is captured with N (θ)xt to allow for the most general case in which the trend depends on the deep parameters. T and with initial values y1 and P1 given. is that the equations are non linear in the deep parameters. As a reminder. Indeed.2 Estimating the likelihood function of the DSGE model The next logical step is to estimate the likelihood of the DSGE solution system mentioned above. This is exactly what Dynare does. chapter 13. are the priors. they are linear in the endogenous and exogenous variables so that the likelihood may be evaluated with a linear prediction error algorithm like the Kalman filter. and are also known. it’s not a coincidence!). . The first apparent problem. The first and second equations above therefore naturally make up a system of measurement and transition or state equations.CHAPTER 8. the recursion follows ∗ vt = yt − y ∗ − M yt − N xt ¯ ˆ Ft = M Pt M + V Kt = gy Pt gy Ft−1 yt+1 = gy yt + Kt vt ˆ ˆ Pt+1 = gy Pt (gy − Kt M ) + gu Qgu For more details on the Kalman filter. The second. it is possible to derive the log-likelihood given by T Tk 1 1 ∗ ln L (θ|YT ) = − ln(2π) − |Ft | − vt Ft−1 vt 2 2 2 t=1 where the vector θ contains the parameters we have to estimate: θ. as is typical for a Kalman filter (you guessed it. it may have a trend. . Yet. . recall. V (θ) and ∗ ∗ Q(θ) and where YT expresses the set of observable endogenous variables yt found in the measurement equation. For t = 1. here’s what the Kalman filter recursion does. see Hamilton (1994). . the log posterior kernel can be expressed as ∗ ∗ ln K(θ|YT ) = ln L (θ|YT ) + ln p(θ) where the first term on the right hand side is now known after carrying out the Kalman filter recursion. respectively.

8. We resort. to sampling-like methods.3. it is a nonlinear and complicated function of the deep parameters θ. but fortunately doable with modern computers. where this is typically the posterior mode. the Metropolis-Hastings algorithm implements the following steps: 1. to find the mode of the posterior distribution .4 Estimating the posterior distribution Finally. p. cΣm ) where Σm is the inverse of the Hessian computed at the posterior mode. instead.a key parameter and an important output of Dynare . This is indeed the method adopted by Dynare. . “constructs a Gaussian approximation around the posterior mode and uses a scaled version of the asymptotic covariance matrix as the covariance matrix for the proposal distribution. Choose a starting point θ ◦ . The general idea of the Metropolis-Hastings algorithm is to simulate the posterior distribution. we cannot obtain an explicit form for it.8. More precisely. To do so. Remember that all we have is the posterior mode. Thus. The distribution will be given by the kernel equation above. Thus. we are instead more often interested in the mean and variance of the estimators of θ. but again. of which the Metropolis-Hastings has been retained in the literature as particularly efficient. This is done in Dynare using numerical methods.3.3. this maximization problem is not completely straightforward.we simply maximize the above log posterior kernel with respect to θ. 18). The algorithm. Recall that the likelihood function is not Gaussian with respect to θ but to functions of θ as they appear in the state equation. 2. in the words of An and Shorfheide. we are now in a position to find the posterior distribution of our parameters. DSGE MODELS AND BAYESIAN ESTIMATION 87 8. This allows for an efficient exploration of the posterior distribution at least in the neighborhood of the mode” (An and Schorfheide (2006). the algorithm builds on the fact that under general conditions the distribution of the deep parameters will be asymptotically normal.3 Finding the mode of the posterior distribution Next. It is a “rejection sampling algorithm” used to generate a sequence of samples (also known as a “Markov Chain” for reasons that will become apparent later) from a distribution that is unknown at the outset. and run a loop over 2-3-4. Draw a proposal θ ∗ from a jumping distribution J(θ ∗ |θ t−1 ) = N (θ t−1 .

go back to the candidate of last period (this is true in very coarse terms. ESTIMATING DSGE MODELS . If the scale factor is too small.4 tries to clarify the above. do two things. Compute the acceptance ratio r= K(θ ∗ |YT ) p(θ ∗ |YT ) = p(θ t−1 |YT ) K(θ t−1 |YT ) 4. the jumping distribution: θt = θ∗ θ t−1 with probability min(r. Metaphorically. After having repeated these steps often enough. and update. if the scale factor is too large. an important parameter in this searching procedure is the variance of the jumping distribution and in particular the scale factor. whose mean has been set to θt−1 (this will become clear in just a moment). 1) otherwise. then definitely keep your candidate. Then. Finally accept or discard the proposal θ ∗ according to the following rule. In step 2. On the other hand. the acceptance rate will be very low (as the candidates are likely to land in regions of low probability density) and the chain will spend too much . Of course. meaning that the distribution will take a long time to converge to the posterior distribution since the chain is likely to get “stuck” around a local maximum.CHAPTER 8. Of course. Update the mean of your drawing distribution. decide whether or not to hold on to your candidate parameter. θ∗ from a Normal distribution. Otherwise. and instead take a few small steps down in the hope of being able to take a big step up in the near future. just in case using that candidate for the mean of the drawing distribution allows us to to leave a local maximum and travel towards the global maximum. build a histogram of those retained values. if necessary. In step 3. notice that in fact you would keep your candidate only with a probability less than one).BEHIND THE SCENES 88 OF DYNARE 3. the point is for each “bucket” of the histogram to shrink to zero. the idea is to allow the search to turn away from taking a small step up. We should not be too quick to simply throw out the candidate giving a lower value of the posterior kernel. Figure 8. choose a candidate paramter. If the acceptance ratio is greater than one. in the final step. and compare it to the value of the kernel from the mean of the drawing distribution. the acceptance rate (the fraction of candidate parameters that are accepted in a window of time) will be too high and the Markov Chain of candidate parameters will “mix slowly”. But why have such a complicated acceptance rule? The point is to be able to visit the entire domain of the posterior distribution. compute the value of the posterior kernel for that candidate parameter. and note the value of the parameter your retain.3. In step 1. This “smoothed histogram” will eventually be the posterior distribution after sufficient iterations of the above steps.

time in the tails of the posterior distribution. DSGE MODELS AND BAYESIAN ESTIMATION 89 Figure 8. These include: How should we choose the scale factor c (variance of the jumping distribution)? What is a satisfactory acceptance rate? How many draws are ideal? How is convergence of the Metropolis-Hastings iterations assessed? These are all important questions that will come up in your usage of Dynare. and smoothing the “histogram” such that each “bucket” has infinitely small width.1: The above sketches the Metropolis-Hastings algorithm.7 of Chapter 5. several practical questions arise when carrying out Bayesian estimation.3. used to build the posterior distribution function. . at least to a machine needing to undertake the above calculations.8. They are addressed as clearly as possible in section 5. While these steps are mathematically clear. Imagine repeating these steps a large number of times.

I)p(θ m |I) I I I where θ m is the posterior mode. which is also the denominator of the posterior density p(θ|YT ) discussed earlier. while touting the advantages of Bayesian estimation. the posterior distribution offers a particularly natural method of comparing models.B p(I)p(YT |I) where this formula may easily be generalized to a collection of N models. where I = A. B we can evaluate. specifying the basic mechanics of Bayesian estimation. B p(I|YT ) = p(I)p(YT |I) I=A. we can compute the posterior distribution over models.BEHIND THE SCENES 90 OF DYNARE 8. The first is to assume a functional form of the posterior kernel that we can integrate. For each model I = A. I)dθ I = ΘI p(θ I |I) × p(YT |θ I . To remind you of this. the comparison of the two models is done very naturally through the ratio of the posterior model distributions. Let’s look at an illustration. Using Bayes’ rule.4 Comparing models using posterior distributions As mentioned earlier. I)dθ I Note that the expression inside the integral sign is exactly the posterior kernel. In this case. Suppose we have a prior distribution over two competing models: p(A) and p(B).CHAPTER 8. we would have the following estimator: p(YT |I) = (2π) 2 |Σθm | 2 p(θ m |YT . p(YT |I). the marginal density of the data conditional on the model by integrating out the deep parameters θ I from the posterior kernel: p(YT |I) = ΘI p(θ I . The most straightforward and the best approximation. The advantage of this technique is its comI putational efficiency: time consuming Metropolis-Hastings iterations are not k 1 . To obtain the marginal density of the data conditional on the model. especially for large samples. This requires some detailed explanations of their own. is the Gaussian (called a Laplace approximation). there are two options. We call this the posterior odds ratio: p(A|YT ) p(A) p(YT |A) = p(B|YT ) p(B) p(YT |B) The only complication is finding the magrinal density of the data conditional on the model. YT |θ I . Then. ESTIMATING DSGE MODELS . you may want to glance back at the first subsection above. at least theoretically.

8.4. COMPARING MODELS USING POSTERIOR DISTRIBUTIONS 91 necessary, only the numerically calculated posterior mode is required. The second option is instead to use information from the MetropolisHastings runs and is typically referred to as the Harmonic Mean Estimator. The idea is to simulate the marginal density of interest and to simply take an average of these simulated values. To start, note that p(YT |I) = E f (θ I ) θI , I p(θ I |I)p(YT |θ I , I)
−1

where f is a probability density function, since E f (θ I ) θI , I = p(θ I |I)p(YT |θ I , I)
ΘI ΘI

f (θ)dθ

p(θ I |I)p(YT |θ I , I)dθ I

and the numerator integrates out to one (seeGeweke (1999) for more details). This suggests the following estimator of the marginal density 1 p(YT |I) = B
(b) B (b) b=1

f (θ I ) p(θ I |I)p(YT |θ I , I)
(b)

(b)

−1

where each drawn vector θ I comes from the Metropolis-Hastings iterations and where the probability density function f can be viewed as a weights on the posterior kernel in order to downplay the importance of extreme values of θ. Geweke (1999) suggests to use a truncated Gaussian function, leading to what is typically referred to as the Modified Harmonic Mean Estimator.

Chapter 9

Optimal policy under commitment

93

.

Thanks for your precious help! 95 . the best is to compile what users have to say! Please let me know what your most common problem is with Dynare. how Dynare tells you about it and how you solve it.Chapter 10 Troubleshooting To make sure this section is as user friendly as possible.

.

J. 97 . 1–126.” Journal of Economic Literature. 17(2-3). (2004): “A method for taking models to the data. 123(1). and S.” Journal of Economic Dynamics and Control.. 18(1).K. A practical guide. F. Oxford University Press.” CEPREMAP mimeo. M. Juillard (2001a): “Accuracy of stochastic perturbation methods: The case of asset pricing models.” Computational Economics. N. Ireland. 979–999. F.” Econometric Review. (1994): Time Series Analysis. (2001b): “A Higher-Order Taylor Expansion Approach to Simulation of Stochastic Forward-Looking Models with an Application to a Nonlinear Phillips Curve Model. 153–187.. Hamilton. Development and Communication. Koopman (2001): Time Series Analysis by State Space Methods. and M. Juillard. NJ. Collard. XXXVII. R.” Econometric Review. S. Princeton University Press. and F. (2003): “Stochastic simulations with DYNARE. J. 1205–1226. Gali.ference. (1996): “Dynare : a program for the resolution and simulation of dynamic models with forward variables through the use of a relaxation algorithm. Fernandez-Villaverde. P.. 25(6-7)..” Journal of Economic Dynamics and Control.” CEPREMAP working papers 9602. D. Durbin. and J. Schorfheide (2006): “Bayesian Analysis of DSGE Models. CEPREMAP. J. Clarida.. and M. 1661–1707. U. Princeton..” Journal of Econometrics. 125–39. Forthcoming. (1999): “Using Simulation Methods for Bayesian Econometric Models: In. Rubio-Ramirez (2004): “Comparing dynamic equilibrium models to data: a Bayesian approach. J. Gertler (1999): “The Science of Monetary Policy: A New Keynesian Perspective.Bibliography An. 28(6). J. Oxford. Geweke.

. 645–670. F. Nason. (2005): “A Bayesian Look at New Open Economy Macroeconomics. 15(6). Schmitt-Grohe. J. Zellner... and M. The Johns Hopkins University. and F.. . and R. and J.” Economics Working Paper Archive 505. The Johns Hopkins University. P.” Journal of Applied Econometrics. F. Schorfheide (2003): “Do Central Banks Respond to Exchange Rate Movements? A Structural Investigation. 1151–1166. Schorfheide. 9(S). A. 52(6). New York. T. (1971): An Introduction to Bayesian Inference in Econometrics. Rabanal. 1(5). and T.. S. F.Department of Economics. S37–70. Inc. Cogley (1994): “Testing the Implications of LongRun Neutrality for Monetary Business Cycle Models. 28(4).” Journal of Applied Econometrics. Smets.Department of Economics. Wouters (2003): “An Estimated Dynamic Stochastic General Equilibrium Model of the Euro Area.98 BIBLIOGRAPHY Lubik. Uribe (2004): “Solving dynamic general equilibrium models using a second-order approximation to the policy function.” Journal of Monetary Economics. John Wiley & Sons.” Journal of the European Economic Association. Rubio-Ramirez (2005): “Comparing New Keynesian models of the business cycle: A Bayesian approach.” Journal of Economic Dynamics and Control. 755–775.. 1123–1175.” Economics Working Paper Archive 521. M. (2000): “Loss function-based evaluation of DSGE models.

Sign up to vote on this title
UsefulNot useful