Professional Documents
Culture Documents
Panel Data Econometrics: Master of Science in Economics - University of Geneva
Panel Data Econometrics: Master of Science in Economics - University of Geneva
University of Orlans
January 2010
General introduction
Denition
A longitudinal, or panel, data set is one that follows a given sample
of individuals over time, and thus provides multiple observations on
each individual in the sample. (Hsiao,2003, page 2).
General introduction
General introduction
Denitions
A micro-panel data set is a panel for which the time dimension T
is largely less important than the individual dimension N (example:
the University of Michigans Panel Study of Income Dynamics,
PSID with 15,000 individuals observed since 1968):
T << N
T 'N
General introduction
Denition
A panel is said to be balanced if we have the same time periods,
t = 1, .., T , for each cross section observation. For an unbalanced
panel, the time dimension, denoted Ti , is specic to each
individual.
Remark: While the mechanics of the unbalanced case are similar
to the balanced case, a careful treatment of the unbalanced case
requires a formal description of why the panel may be unbalanced,
and the sample selection issues can be somewhat subtle. =>
issues of sample selection and attrition.
General introduction
General introduction
General introduction
What are the main advantages of the panel data sets and
the panel data models?
General introduction
What are the main advantages of the panel data sets and
the panel data models?
Denition
The oft-touted power of panel data derives from their theoretical
ability to isolate the eects of specic actions, treatments, or more
general policies.
General introduction
General introduction
What are the main advantages of the panel data sets and
the panel data models?
General introduction
Example
Example: let us consider a simple regression model.
where
xit and zit are k1 1 and k2 1 vectors of exogeneous
variables
is a constant, and are k1 1 and k2 1 vectors of
parameters
it is i.i.d. over i and t, with V (it ) = 2
Let us assume that zit variables unobservable and
correlated with xit
cov (xit , zit ) 6= 0
C. Hurlin Panel Data Econometrics
Denitions
What are the main advantages of the panel data sets and the panel d
General introduction
Issues involved in utilizing panel data
Outline of the lecture
General introduction
yit yi ,t 1 = 0 (xit xi ,t 1 ) + it i ,t 1
General introduction
y t = (1/N ) N
i =1 yit x t = (1/N ) N
i =1 xit t = (1/N ) N
i =1 it
General introduction
What are the main advantages of the panel data sets and
the panel data models?
General introduction
xt = xt 1 + t
d 1 W (1)2 1
T (b
1) ! R
T ! 2 1 W (r )2 dr
0
General introduction
General introduction
xi ,t = xi ,t 1 + i ,t
General introduction
General introduction
General introduction
General introduction
General introduction
Example
Let us consider the example of a simple production function (Cobb
Douglas) with two factors (labor and capital). We have N
countries and T periods. Let us denote:
- yit the log of the GDP for country i at time t.
- nit the log of the labor employment for country i at time t.
- yit the log of the capital stock for country i at time t.
yi ,t = i + i ki ,t + i ni ,t + i ,t
with i ,t i.i.d. 0, 2 , 8 i, 8 t.
General introduction
Example
In this specication, the elasticities i and i are specic to each
country
yi ,t = i + i ki ,t + i ni ,t + i ,t
with i ,t i.i.d. 0, 2 , 8 i, 8 t. But, several alternative
specications can be considered. First, we can assume that the
production function is the same for all countries: in this case we
have an homogeneous specication:
yi ,t = + ki ,t + ni ,t + i ,t
i = i = i =
Example
However, an homogeneous specication of the production function
for macro aggregated data is meaningless. We can introduce an
heterogeneity of the Total Factor Productivity: more precisely, we
can assume that the mean of TFP (given by E (i + i ,t ) = i ) is
dierent accross countries (due to institutional organisational
factors, etc.).
Then, we can use a specication with individual eects, i and
common slope parameters (elasticities and ).
yi ,t = i + ki ,t + ni ,t + i ,t
i = i =
Example
Finally, we can assume that the labor and/or capital elasticities are
dierent accross countries.In this case, we will have an
heterogeneous specication of the panel data model
(heterogeneous panel).
yi ,t = i + i ki ,t + i ni ,t + i ,t
Example
In this case, there are two solutions:
1 ) The rst solution consists in using N times series models to
produce some group-mean estimates of the elasticities.
2 ) The second solution consists in using a model with random
(slope) parameters => random coe cient model. In this
case, we assume that parameters i and i and randomly
distributed, but follows the same distribution:
i i.i.i , 2 i i.i.i , 2
yit = i + xit + it
yit = + xit + it
General introduction
General introduction
General introduction
yit = i + i xit + it
yit = + xit + it
General introduction
General introduction
In this case, pooling gives rise to the false inference that the
pooled relation is curvilinear.
Fact
In both cases, the classic paradigm of the representative agent
simply does not hold, and pooling the data under homogeneity
assumption makes no sense.
General introduction
General introduction
General introduction
References
Articles (chapter 2)
Baltagi, B.H. et Kao, C. (2000), Nonstationary panels,
cointegration in panels and dynamic panels : a survey, in
Advances in Econometrics, 15, edited by B. Baltagi et C. Kao,
pp. 7-51, Elsevier Science.
Hurlin, C. et Mignon, V. (2005), Une synthse des tests de
racine unitaire sur donnes de panel, Economie et Prvision,
169-170-171, pp. 253-294
http://www.univ-orleans.fr/deg/masters/ESA/CH/churlin_E.htm