Chapter-11Panel Data

Introductory Econometrics for Finance
Chapter 11
Panel Data
‘Introductory Econometrics for Finance’ © Chris Brooks 2019 1
Learning Outcomes
• Describe the key features of panel data and outline the advantages and
disadvantages of working with panels rather than other structures
• Explain the intuition behind seemingly unrelated regressions
• Contrast the fixed effect and random effect approaches to panel model
specification, determining which is the more appropriate in particular
cases
Copyright 2019 by Chris Brooks 1

The Nature of Panel Data
• Panel data, also known as longitudinal data, have both time series and cross-
sectional dimensions.
• They arise when we measure the same collection of people or objects over a
period of time.
• Econometrically, the setup is
yit    xit  uit
where yit is the dependent variable,  is the intercept term,  is a k  1 vector
of parameters to be estimated on the explanatory variables, xit; t = 1, …, T;
i = 1, …, N.
• The simplest way to deal with this data would be to estimate a single, pooled
regression on all the observations together.
• But pooling the data assumes that there is no heterogeneity – i.e. the same
relationship holds for all the data.
The Advantages of using Panel Data
• There are a number of advantages from using a full panel technique when a
panel of data is available.
– We can address a broader range of issues and tackle more complex

problems with panel data than would be possible with pure time series or
pure cross-sectional data alone.
– It is often of interest to examine how variables, or the relationships
between them, change dynamically (over time). Can also help to
mitigate problems of multicollinearity that may arise if time series are
modelled individually
– By structuring the model in an appropriate way, we can remove the
impact of certain forms of omitted variables bias in regression results.

What panel techniques are available?
Seemingly Unrelated Regression (SUR)
• One approach to making more full use of the structure of the data would be to use the
SUR framework initially proposed by Zellner (1962). This has been used widely in
finance where the requirement is to model several closely related variables over time.
• A SUR is so-called because the dependent variables may seem unrelated across the
equations at first sight, but a more careful consideration would allow us to conclude
that they are in fact related after all.
• Under the SUR approach, one would allow for the contemporaneous relationships
between the error terms in the equations by using a generalised least squares (GLS)
technique.
• The idea behind SUR is essentially to transform the model so that the error terms
become uncorrelated.
• If the correlations between the error terms in the individual equations had been zero
in the first place, then SUR on the system of equations would have been equivalent to
running separate OLS regressions on each equation.

Fixed and Random Effects Panel Estimators
• The applicability of the SUR technique is limited because it can only be

employed when the number of time series observations (T) per cross-
sectional unit i is at least as large as the total number of such units, N.
• A second problem with SUR is that the number of parameters to be estimated

in total is very large, and the variance-covariance matrix of the errors also
has to be estimated (NT x NT). For these reasons, the more flexible full panel
data approach is much more commonly used.
• There are two main classes of panel techniques:

– the fixed effects estimator and
– the random effects estimator.
Fixed Effects Models
• The fixed effects model for some variable yit may be written
yit    xit  i  vit
• We can think of i as encapsulating all of the variables that affect yit cross-
sectionally but do not vary over time – for example, the sector that a firm
operates in, a person's gender, or the country where a bank has its
headquarters, etc. Thus we would capture the heterogeneity that is
encapsulated in i by a method that allows for different intercepts for each
cross sectional unit.
• This model could be estimated using dummy variables, which would be

termed the least squares dummy variable (LSDV) approach.

Fixed Effects Models (Cont’d)
• The LSDV model may be written
yit    xit  1D1i  2 D2i  3 D3i     N DNi  vit

where D1i is a dummy variable that takes the value 1 for all observations on the first
entity (e.g., the first firm) in the sample and zero otherwise, D2i is a dummy variable
that takes the value 1 for all observations on the second entity (e.g., the second firm)
and zero otherwise, and so on.
• The LSDV can be seen as just a standard regression model and therefore it can be
estimated using OLS.
• Now the model given by the equation above has N+k parameters to estimate.
• In order to avoid the necessity to estimate so many dummy variable parameters, a

transformation, known as the within transformation, is used to simplify matters.
The Within Transformation
• The within transformation involves subtracting the time-mean of each entity

away from the values of the variable.
• So define yi  Tt1 yit as the time-mean of the observations for cross-sectional
unit i, and similarly calculate the means of all of the explanatory variables.
• Then we can subtract the time-means from each variable to obtain a
regression containing demeaned variables only.
• Note that such a regression does not require an intercept term since now the
dependent variable will have zero mean by construction.
• The model containing the demeaned variables is yit  yi   ( xit  xi )  uit  ui
• We could write this as yit  xit 
 uit , where the double dots above the
variables denote the demeaned values.
• This model can be estimated using OLS, but we need to make a degrees of
freedom correction.

The Between Estimator
• An alternative to this demeaning would be to simply run a cross-sectional

regression on the time-averaged values of the variables, which is known as
the between estimator.
• An advantage of running the regression on average values (the between
estimator) over running it on the demeaned values (the within estimator) is
that the process of averaging is likely to reduce the effect of measurement
error in the variables on the estimation process.
• A further possibility is that instead, the first difference operator could be
applied so that the model becomes one for explaining the change in yit rather
than its level. When differences are taken, any variables that do not change
over time will again cancel out.
• Differencing and the within transformation will produce identical estimates
in situations where there are only two time periods.
Time Fixed Effects Models
• It is also possible to have a time-fixed effects model rather than an entity-

fixed effects model.
• We would use such a model where we think that the average value of yit
changes over time but not cross-sectionally.
• Hence with time-fixed effects, the intercepts would be allowed to vary over
time but would be assumed to be the same across entities at each given point
in time. We could write a time-fixed effects model as
yit    xit   t  vit
where t is a time-varying intercept that captures all of the variables that
affect y and that vary over time but are constant cross-sectionally.
– An example would be where the regulatory environment or tax rate changes part-way
through a sample period. In such circumstances, this change of environment may well
influence y, but in the same way for all firms.

Time Fixed Effects Models (Cont’d)
• Time-variation in the intercept terms can be allowed for in exactly the same way as
with entity fixed effects. That is, a least squares dummy variable model could be
estimated
yit  xit  1D1t  2 D 2t  ...  T DTt  vit
where D1t, for example, denotes a dummy variable that takes the value 1 for the first
time period and zero elsewhere, and so on.
• The only difference is that now, the dummy variables capture time variation rather
than cross-sectional variation.
• Similarly, in order to avoid estimating a model containing all T dummies, a within
transformation can be conducted to subtract away the cross-sectional averages from
each observation
• Finally, it is possible to allow for both entity fixed effects and time fixed effects
within the same model. Such a model would be termed a two-way error component
model, and the LSDV equivalent model would contain both cross-sectional and time
dummies
Results from Test of Banking Market Equilibrium

by Matthews et al.
• The null hypothesis that the bank fixed effects are jointly zero (H0: i = 0) is rejected at the 1%
significance level for the full sample and for the second sub-sample but not at all for the first sub-
sample. Overall, however, this indicates the usefulness of the fixed effects panel model that
allows for bank heterogeneity.
Source: Matthews, Murinde and Zhao (2007)


The Random Effects Model
• An alternative to the fixed effects model described above is the random

effects model, which is sometimes also known as the error components
model.
• As with fixed effects, the random effects approach proposes different
intercept terms for each entity and again these intercepts are constant over
time, with the relationships between the explanatory and explained variables
assumed to be the same both cross-sectionally and temporally.
• However, the difference is that under the random effects model, the
intercepts for each cross-sectional unit are assumed to arise from a common
intercept  (which is the same for all cross-sectional units and over time),
plus a random variable i that varies cross-sectionally but is constant over
time.
• i measures the random deviation of each entity’s intercept term from the
“global” intercept term . We can write the random effects panel model as
yit    xit  it , it   i  vit
How the Random Effects Model Works
• Unlike the fixed effects model, there are no dummy variables to capture the
heterogeneity (variation) in the cross-sectional dimension.
• Instead, this occurs via the i terms.
• Note that this framework requires the assumptions that the new cross-sectional error
term, i, has zero mean, is independent of the individual observation error term vit, has
constant variance, and is independent of the explanatory variables.
• The parameters ( and the  vector) are estimated consistently but inefficiently by
OLS, and the conventional formulae would have to be modified as a result of the
cross-correlations between error terms for a given cross-sectional unit at different
points in time.
• Instead, a generalised least squares (GLS) procedure is usually used. The

transformation involved in this GLS procedure is to subtract a weighted mean of the
yit over time (i.e. part of the mean rather than the whole mean, as was the case for
fixed effects estimation).

Quasi-Demeaning the Data
• Define the ‘quasi-demeaned’ data as yit*  yit  yi and similarly for xit,
•  will be a function of the variance of the observation error term, v2, and of
the variance of the entity-specific error term, 2:
v
 1
T 2   v2
• This transformation will be precisely that required to ensure that there are no
cross-correlations in the error terms, but fortunately it should automatically
be implemented by standard software packages.
• Just as for the fixed effects model, with random effects, it is also
conceptually no more difficult to allow for time variation than it is to allow
for cross-sectional variation.
Quasi-Demeaning the Data
• In the case of time-variation, a time period-specific error term is included
• Again, a two-way model could be envisaged to allow the intercepts to vary

both cross-sectionally and over time.

Fixed or Random Effects?
• It is often said that the random effects model is more appropriate when the
entities in the sample can be thought of as having been randomly selected from
the population, but a fixed effect model is more plausible when the entities in the
sample effectively constitute the entire population.
• More technically, the transformation involved in the GLS procedure under the
random effects approach will not remove the explanatory variables that do not
vary over time, and hence their impact can be enumerated.
• Also, since there are fewer parameters to be estimated with the random effects
model (no dummy variables or within transform to perform), and therefore
degrees of freedom are saved, the random effects model should produce more
efficient estimation than the fixed effects approach.
• However, the random effects approach has a major drawback which arises from
the fact that it is valid only when the composite error term it is uncorrelated with
all of the explanatory variables.
Fixed or Random Effects? (Cont’d)
• This assumption is more stringent than the corresponding one in the fixed effects
case, because with random effects we thus require both i and vit to be independent of
all of the xit.
• This can also be viewed as a consideration of whether any unobserved omitted

variables (that were allowed for by having different intercepts for each entity) are
uncorrelated with the included explanatory variables. If they are uncorrelated, a
random effects approach can be used; otherwise the fixed effects model is preferable.
• A test for whether this assumption is valid for the random effects estimator is based
on a slightly more complex version of the Hausman test.
• If the assumption does not hold, the parameter estimates will be biased and
inconsistent.
• To see how this arises, suppose that we have only one explanatory variable, x2it that
varies positively with yit, and also with the error term, it. The estimator will ascribe
all of any increase in y to x when in reality some of it arises from the error term,
resulting in biased coefficients.

Credit Stability of Banks in Central and Eastern

Europe: A Random Effects Analysis
• Foreign participants in the banking sector may improve competition and
efficiency to the benefit of the economy that they enter.
• They may have a stabilising effect on credit provision since they will
probably be better diversified than domestic banks and will therefore be more
able to continue to lend when the host economy is performing poorly.
• But on the other hand, it is also argued that foreign banks may alter the credit
supply to suit their own aims rather than that of the host economy.
• They may act more pro-cyclically than local banks, since they have
alternative markets to withdraw their credit supply to when host market
activity falls.
• Moreover, worsening conditions in the home country may force the
repatriation of funds to support a weakened parent bank.
The Data
• There may also be differences in policies for credit provision dependent upon
the nature of the formation of the subsidiary abroad – i.e. whether the
subsidiary's existence results from a take-over of a domestic bank or from the
formation of an entirely new startup operation (a ‘greenfield investment’).
• A study by de Haas and van Lelyveld (2006) employs a panel regression

using a sample of around 250 banks from ten Central and East European
countries.
• They examine whether domestic and foreign banks react differently to

changes in home or host economic activity and banking crises.
• The data cover the period 1993-2000 and are obtained from BankScope.

The Model
• The core model is a random effects panel regression of the form:

grit    1Takeoverit  2Greenfieldi   3Crisisit   4 Macroit
 5Contrit  ( i   it )
where the dependent variable, grit, is the percentage growth in the credit of
bank i in year t; Takeover is a dummy variable taking the value one for
foreign banks resulting from a takeover and zero otherwise; Greenfield is a
dummy taking the value one if bank is the result of a foreign firm making a
new banking investment rather than taking over an existing one; crisis is a
dummy variable taking the value one if the host country for bank i was
subject to a banking disaster in year t.
• Macro is a vector of variables capturing the macroeconomic conditions in the
home country (the lending rate and the change in GDP for the home and host
countries, the host country inflation rate, and the differences in the home and
host country GDP growth rates and the differences in the home and host
country lending rates).
The Model (Cont’d)
• Contr is a vector of bank-specific control variables that may affect the

dependent variable irrespective of whether it is a foreign or domestic bank.
• These are: weakness parent bank, defined as loan loss provisions made by the
parent bank; solvency is the ratio of equity to total assets; liquidity is the ratio
of liquid assets / total assets; size is the ratio of total bank assets to total
banking assets in the given country; profitability is return on assets and
efficiency is net interest margin.
•  and the 's are parameters (or vectors of parameters in the cases of 4 and
5), i is the unobserved random effect that varies across banks but not over
time, and it is an idiosyncratic error term.

Estimation Options
• de Haas and van Lelyveld discuss the various techniques that could be
employed to estimate such a model.
• OLS is considered to be inappropriate since it does not allow for differences
in average credit market growth rates at the bank level.
• A model allowing for entity-specific effects (i.e. a fixed effects model that
effectively allowed for a different intercept for each bank) is ruled out on the
grounds that there are many more banks than time periods and thus too many
parameters would be required to be estimated.
• They also argue that these bank-specific effects are not of interest to the
problem at hand, which leads them to select the random effects panel model.
• This essentially allows for a different error structure for each bank. A
Hausman test is conducted, and shows that the random effects model is valid
since the bank-specific effects i are found “in most cases not to be
significantly correlated with the explanatory variables.”
Results
Source: de Haas and van Lelyveld (2006)

Chapter-11Panel Data

Uploaded by

Copyright:

Available Formats

Chapter-11Panel Data

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter-11Panel Data

Uploaded by

Copyright:

Available Formats

Introductory Econometrics for Finance

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 1

• Explain the intuition behind seemingly unrelated regressions

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 2

Copyright 2019 by Chris Brooks 1

The Nature of Panel Data

The Advantages of using Panel Data

– We can address a broader range of issues and tackle more complex

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 4

Copyright 2019 by Chris Brooks 2

What panel techniques are available?

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 5

Seemingly Unrelated Regression (SUR)

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 6

Copyright 2019 by Chris Brooks 3

Fixed and Random Effects Panel Estimators

• The applicability of the SUR technique is limited because it can only be

• A second problem with SUR is that the number of parameters to be estimated

• There are two main classes of panel techniques:

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 7

Fixed Effects Models

• This model could be estimated using dummy variables, which would be

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 8

Copyright 2019 by Chris Brooks 4

Fixed Effects Models (Cont’d)

• The LSDV model may be written

yit    xit  1D1i  2 D2i  3 D3i     N DNi  vit

• In order to avoid the necessity to estimate so many dummy variable parameters, a

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 9

The Within Transformation

• The within transformation involves subtracting the time-mean of each entity

Copyright 2019 by Chris Brooks 5

The Between Estimator

• An alternative to this demeaning would be to simply run a cross-sectional

Time Fixed Effects Models

• It is also possible to have a time-fixed effects model rather than an entity-

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 12

Copyright 2019 by Chris Brooks 6

Time Fixed Effects Models (Cont’d)

Results from Test of Banking Market Equilibrium

Source: Matthews, Murinde and Zhao (2007)

Copyright 2019 by Chris Brooks 7

The Random Effects Model

• An alternative to the fixed effects model described above is the random

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 23

How the Random Effects Model Works

• Instead, a generalised least squares (GLS) procedure is usually used. The

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 24

Copyright 2019 by Chris Brooks 8

Quasi-Demeaning the Data

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 25

Quasi-Demeaning the Data

• In the case of time-variation, a time period-specific error term is included

• Again, a two-way model could be envisaged to allow the intercepts to vary

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 26

Copyright 2019 by Chris Brooks 9

Fixed or Random Effects?

‘Introductory Econometrics for Finance’ © Chris Brooks 2019 27

Fixed or Random Effects? (Cont’d)

• This can also be viewed as a consideration of whether any unobserved omitted