Chapter 14

Advanced Panel
Data Methods

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or
service or otherwise on a password-protected website or school-approved learning management system for classroom use. © kentoh/Shutterstock.
 Advanced Panel Data Methods

● Fixed effects estimation

Fixed effect, potentially corre-
lated with explanatory variables

Form time-averages
for each individual

Because (the fixed effect is removed)

● Estimate time-demeaned equation by OLS

• Uses time variation within cross-sectional units (= within estimator)

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Example: Effect of training grants on firm scrap rate

Time-invariant reasons why one firm is more productive than another are controlled
for. The important point is that these may be correlated with the other explanat.
variables.
Stars denote
Fixed-effects estimation using the years 1987, 1988, and 1989:
time-demeaning

Training grants significantly improve productivity (with a time lag)

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Discussion of fixed effects estimator

• Strict exogeneity in the original model has to be assumed
• The R-squared of the demeaned equation is inappropriate
• The effect of time-invariant variables cannot be estimated
• But the effect of interactions with time-invariant variables can be
estimated (e.g. the interaction of education with time dummies)
• If a full set of time dummies are included, the effect of variables
whose change over time is constant cannot be estimated (e.g.
experience)
• Degrees of freedom have to be adjusted because the N time
averages are estimated in addition (resulting degrees of freedom =
NT-N-k)

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Interpretation of fixed effects as dummy variable regression

• The fixed effects estimator is equivalent to introducing a dummy for
each individual in the original regression and using pooled OLS:

For example, = 1 if the observation

stems from individual N, = 0 otherwise

• After fixed effects estimation, the fixed effects can be estimated as:

Estimated individual
effect for individual i

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Fixed effects or first differencing?

• Remember that first differencing can also be used if T > 2
• In the case T = 2, fixed effects and first differencing are identical
• For T > 2, fixed effects is more efficient if classical assumptions hold
• First differencing may be better in the case of severe serial
correlation in the errors, for example if the errors follow a random
walk
• If T is very large (and N not so large), the panel has a pronounced
time series character and problems such as strong dependence arise
• In these cases, it is probably better to use first differencing
• Otherwise, it is a good idea to compute both and check robustness

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Random effects models The individual effect is assumed to be

“random” i.e. completely unrelated to
explanatory variables

Random effects assumption:

The composite error ai + uit is uncorrelated with the explanatory variables

but it is serially correlated for observations coming from the same i:
Under the assumption that
idiosyncratic errors are
serially uncorrelated

For example, in a wage equation, for a given individual the same unobserved ability appears in
the error term of each period. Error terms are thus correlated across periods for this individual.

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Estimation in the random effects model

• Under the random effects assumptions explanatory variables are
exogenous so that pooled OLS provides consistent estimates
• If OLS is used, standard errors have to be adjusted for the fact that
errors are correlated over time for given i (= clustered standard
errors)
• But, because of the serial correlation, OLS is not efficient
• One can transform the model so that it satisfies the GM-
assumptions:
Quasi-demeaned data

Error can be shown to satisfy GM-assumptions

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Estimation in the random effects model (cont.)

with

• The quasi-demeaning parameter is unknown but it can be estimated

• FGLS using the estimated is called random effects estimation
• If the random effect is relatively unimportant compared to the
idosyn-cratic error, FGLS will be close to pooled OLS (because
)
• If the random effect is relatively important compared to the idiosyn-
cratic term, FGLS will be similar to fixed effects (because )
• Random effects estimation works for time-invariant variables

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Example: Wage equation using panel data

Random effects is used because

many of the variables are time-
invariant. But is the random
effects assumption realistic?

● Random effects or fixed effects?

• In economics, unobserved individual effects are seldomly
uncorrelated with explanatory variables so that fixed effects is more
convincing
© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Correlated random effects approach

(= Correlated Random Effect, CRE)

The individual-specific effect ai is split up into a part that is related to the time-averages of the expla-
natory variables and a part ri that is uncorrelated to the explanatory variables.

The resulting model is an ordinary random effects model with uncorrelated random effect r i but with
the time averages as additional regressors. It turns out that in this model, the resulting estimates for
the explanatory variables are identical to those of the fixed effects estimator.

Advantages: 1) One can test FE vs. (ordinary) RE, 2) One can incorporate time-invariant regressors

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
 Advanced Panel Data Methods

● Applying panel data methods to other data structures

• Panel data methods can be used in other contexts where constant
unobserved effects have to be removed

● Example: Wage equations for twins

Unobserved genetic and family characteristics that do not vary across twins

Equation for twin 1

in family i

Equation for twin 2

in family i
Estimate differenced
equation by OLS

© 2016 Cengage Learning ®. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a
certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.