You are on page 1of 9

Introduction to Econometrics

Ratjomose P. Machema
rp.machema@nul.ls
EC6041: Econometric Theory and Applications

Table of contents

Contents
1 What is Econometrics? 1

2 Types of Econometric Models? 3

3 Steps in Empirical Economic Analysis 4

4 The Structure of Economic Data 7


4.1 Cross-sectional data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4.2 Time Series Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4.3 Independently Pooled Cross Sections . . . . . . . . . . . . . . . . . . . 9
4.4 Panel Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1 What is Econometrics?
Definitions of Econometrics

• Economic theory helps to solve problems by:


– Identifying important variables
– Illustrating interdependencies between economic variables

• While certainly useful, economic theory is not sufficient to address most


real-world economic problems

1
– at best, it only indicates the direction of change between variables
– Real world problems almost always require prediction of the size of
change
∗ e.g. How much will the farm price of beef increase as average con-
sumer income increases by R1000 per month?

Definitions of Econometrics

• The most important job of econometrics is to quantify relationships between


different quantities on the basis of available data and using statistical tech-
niques, and to interpret, use or exploit the resulting outcomes appropriately.
– Therefore, econometrics is the interaction of economic theory, observed
data and statistical methods.

• We can use econometrics to estimate economic relationships, test economic


theories, and evaluate government and business policy.
– For example, what if we are asked to study the effects of agricultural
spending on farm performance?
∗ A simple correlation analysis might not be sufficient because caus-
ality can be difficult to infer.
∗ (On average, farms that have more revenue per plot also have more
productivity per crop.)

Definitions of Econometrics

• Econometrics is concerned with model building.


– The “model” typically begins with an observation or a proposition that
∗ movement of one variable “is caused by” movement of another, or “a
variable varies with another ,” or
∗ a relationship between a variable and one or more covariates that are
expected to be related to the variable in question.
∗ individuals’ usage of the health care system depends on, for example,
perceived health status, demographics (e.g., income, age, and education),
and the amount and type of insurance they have.
– An econometric model begins with an idea of some kind of relationship, and
the natural next step is to translate that idea into a set of equations that will
answer interesting questions about the variable of interest.

2
2 Types of Econometric Models?
Definitions of Econometrics
Broadly, econometric models can be classified into a number of categories.
A first class of models describes relationships between present and past.

• Mainly built to get forecasts for future values and the corresponding uncertainty or volatility.

• How does the short-term interest rate depend on its own history?

A second type of model considers relationships btwn economic quantities over a certain time period.

• Gives us information on how (aggregate) economic quantities fluctuate over time in relation to other
quantities.

• What happens to the long-term interest rate if the monetary authority adjusts the short-term one?

Definitions of Econometrics
Third are models that describe relationships between different variables measured at a given point in time for
different units (e.g. households or firms).

• Explain why these units are different or behave differently.

– Analyse to what extent differences in household savings can be attributed to differences in house-
hold income.

• Analyse ‘what if’ questions.

• How much more would a given household, or the average household, save if income were to increase by
1%?

A class that considers relationships between different variables measured for different units over a longer time
span (at least two periods).

• Usually requires panel data, repeated observations over the same units.

• Ideally suited for analysing policy changes on an individual level, provided that it can be assumed that
the structure of the model is constant into the (near) future.

3
Definitions of Econometrics
• The job of econometrics is to specify and quantify these relationships.
• Another job of the econometrician is to judge whether the resulting model
is ‘appropriate’.
– That is, to check whether the assumptions made to motivate the es-
timators (and their properties) are correct, and to check whether the
model can be used for its intended purpose.
• The number of econometric techniques that can be used is numerous, and
their validity often depends crucially upon the validity of the underlying
assumptions.
• Thus our approach in this course is to guide you through this forest of es-
timation and testing procedures, not by describing the beauty of all possible
trees.

Definitions of Econometrics
• An distinction is made between micro-econometrics and macro-econometrics.
– Micro-metrics is characterized by its analysis of cross section and
panel data and by its focus on individual consumers, firms, and micro-
level decision makers.
– Macro-metrics is involved in the analysis of time-series data, usually
of broad aggregates such as price levels, the money supply, exchange
rates, output, investment, economic growth, and so on.

3 Steps in Empirical Economic Analysis


Empirical Analysis
• In economics, theory and empirical analysis are both important.
• Econometric methods are relevant in virtually every branch of applied eco-
nomics.
– They come into play either when we have an economic theory to test
or when we have a relationship in mind that has some importance for
business decisions or policy analysis.
• An empirical analysis uses data to test a theory, estimate an economic
relationship, or determine the effects of a policy or intervention.

4
Steps for a Successful Empirical Study
Step 1: Careful formulation of the question of interest

• Be very precise in posing the question you hope to answer. For example,
– does attending lectures in college lead to better grades (on average)?
∗ If the severity of punishment for certain crimes increases, do crime
rates fall on average?

Step 2: Specify an economic model,

• or at least a conceptual model, to study the phenomenon of interest. Formal


economic modelling (such as utility maximization) is often used, but one can
get by with careful economic reasoning that is less formal.

Example of Econ model

• To study the effects of farmer training on farm productivity - where pro-


ductivity is measured by observed crop yield per hectare -
– we can start with an equation such as

crop yield per hectare = f (educ; exper; training) (1)

– where educ is a measure of education level of the farmer, exper is a


measure of farming experience, and training is a measure of time spent
in the training programme (the variable of most interest).

Step 3: Form an Econometric model


Step 1: Carefully pose a question.
Step 2: Specify an economic or conceptual model.
Step 3: Turn the economic model into an econometric model.

• Our focus is on econometric models. Here is where we resolve certain diffi-


culties and ambiguities concerning an economic model. For example,
– What is the exact functional relationship among economic variables?
∗ How do we account for unobserved factors that make relationships
among variables inexact?

5
Example of metrics model
We can specify an econometric model for the crop yield/farming training example as

crop yield per hactare = β0 + β1 educ + β2 exper + β3 training + µ (2)

The constants β0 , β1 , β2 , andβ3 are the parameters of the econometric model, and it is these
(especially β3 in this example) that we hope to estimate.

• Ideally we will be able to collect data on crop yield, educ, exper, and training from a large
group of farmers.

• The last term in the equation (µ) called the error term or disturbance. It plays a very
important role in econometrics.
– It represents all other factors that can affect farmer’s crop yield, intelligence, mo-
tivation, and so on.
– The error term can also capture measurement problems in one or more of the vari-
ables.
– We could add soil quality, climate and any other variables that determine crop yield,
but we can never eliminate µ entirely.
– In fact, dealing with this error term or disturbance term is perhaps the most im-
portant component of any econometric analysis.

Hypotheses

• Once an econometric model such as (2) has been specified, various hypotheses
of interest can be stated in terms of the unknown parameters.

• We will want to use statistical methods, and data, to estimate and test
hypotheses about the parameters.

• For example,
– The hypothesis that farmer training has no effect on crop yield is β3 = 0.
– The hypothesis that one year of farming experience is worth one year
of formal education is β1 = β2 .

Step 4: Collect data on the variables

• An empirical analysis requires data. After data on the relevant variables


have been collected,
– econometric methods are used to estimate the parameters in the eco-
nometric model and to formally test hypotheses of interest.
– Because we use data to answer economic questions, our answers always
have some uncertainty:

6
∗ A different set of data would produce a different answer.Therefore,
the conceptual framework for the empirical analysis needs to provide
both an answer to the question and a measure of how precise the
answer is. That is, construct confidence intervals for the paramet-
ers.
– In some cases, the econometric model is used to make predictions in
either the testing of a theory or the study of a policy’s impact.

4 The Structure of Economic Data


Economic Data
Data may be collected at various levels of aggregation:
Micro

• data collected on individual economic decision-making units such as individuals, house-


holds, and firms.

Macro

• data resulting from a pooling or aggregating over individuals, households, or firms at the
local, state, or national levels.

Data may be quantitative or qualitative:


Quantitative

• outcomes such as prices or income that may be expressed as numbers or some transfor-
mation of them, such as real prices or per capita income.

Qualitative
• outcomes that are of an “either-or” situation. For example, a consumer either did or did
not make a purchase of a particular good, or a person either is or is not married.

Economic Data

• Economic data come in several different forms.


– Cross-sectional data
– Time series data.
– Pooled cross sections and panel data structures

7
4.1 Cross-sectional data
Cross-sectional data
• Cross-sectional data set are collected on individuals/farmers, families, firms, farms, or
some other units at a given point in time.
– (Or, at least, time does not play a crucial role. Interview dates for surveys may vary
somewhat.)
– For example, several families may be surveyed during different weeks within a year.
• In this course, we will assume that a cross-sectional data set represents a random sample.
– That is, each unit in the population has the same chance of appearing the sample,
and the draws are statistically independent of one another.
– For example, if we obtain information on wages, education, experience, and other
characteristics by randomly drawing 500 people from the working population, then
we have a random sample from the population of all working people.

With cross-sectional data, we can learn about relationships among variables by studying differences
across people, firms, or other economic entities during a single time period.

4.2 Time Series Data


Time Series Data
• Time series data are data for a single entity (person, firm, country) collected at multiple
time periods.

• Consists of observations on variables observed over a stretch of time.


– Examples include interest rates, unemployment rates, and crime rates.
• A key feature of time series data is that the order is (usually) important.
– We need to know, for example, that the outcome on unemployment in 2008 precedes
that for 2009.

• Another important difference with cross-sectional data is that we cannot assume outcomes
are independent across observation (that is, across time).
– For example, knowing what gross domestic product is in 2009 tells us a lot about
its likely range in 2010.
– When we apply econometric methods to time series data, we will have to recog-
nize that the observations are correlated across time. Sometimes this correlation is
substantial.

By tracking a single entity over time, time series data can be used to study the evolution of
variables over time and to forecast future values of those variables.

8
4.3 Independently Pooled Cross Sections
Pooled Cross Sections

• A data set consisting of independently pooled cross sections means that we have
collect cross-sectional data at different points in time and pool them together.
– For example, we may randomly sample from the producers of crop/grains in the
country 1990, 2000, and 2010.
– Our goal may be to see how the importance of input subsidies on productivity has
changed over time.

• If we obtain a random sample in each year it would be very small compared to the
entire population.
– It would be very rare that the same person would appear twice; if someone appears
twice nothing is harmed by ignoring that fact.

• Pooled cross sections are very useful for policy analysis - to study an intervention.
• Because the observations are independent (both within and across time periods),
pooled cross sections can be analyzed much like a single cross section.
– However, one often explicitly accounts for the different time periods in analyzing
the data.

4.4 Panel Data


Panel Data

• Superficially, a panel data set has a structure similar to a pooled cross section.

• The key difference is that with a panel data set, the same units (people,
houses, schools, and so on) are followed over time.

• Panel data, also called longitudinal data, are data for multiple entities
in which each entity is observed at two or more time periods.

• For example, there are a couple of poultry farms in Lesotho.


– We can collect information on output, spending, and some socio-economic
variables annually over, say, the last 10 years.

• Following the same units over time has advantages when trying to infer
causality.

You might also like