[Gretl-users] Difference between pooled and panel data

John C Frain frainj at gmail.com
Tue Jul 12 16:29:03 EDT 2011

To me the term pooled refers to a type of model.  For example in a
fixed effects model is equivalent to having a distinct model for each
level of the factor that determines the panel.  Pooled OLS corresponds
to pooling the data set over all levels of the factor and estimating a
single OLS (assuming that the fixed effects are constant over all
levels of the factor. With a multilevel model pooling corresponds to
collapsing some levels of the  model to get a pooled model.  When some
individuals leave the panel and others join I would call this an
unbalanced panel.

Best Regards


On 12 July 2011 13:26, Allin Cottrell <cottrell at wfu.edu> wrote:
> On Tue, 12 Jul 2011, Anutechia Asongu wrote:
>>         Could someone remind me the difference between pooled and panel
>> data.
>>         Cheers and thanks
> Panel data are data that track measures on a given, fixed set of
> "individuals" over time. The phrase "pooled data" is more general: it
> implies that the data have both a time-series and a cross-sectional
> dimension, but it includes the case where the individuals are not a fixed
> set -- e.g. some individuals drop out over time and are replaced by others.
> At least that's how I read it.
> Allin Cottrell
> _______________________________________________
> Gretl-users mailing list
> Gretl-users at lists.wfu.edu
> http://lists.wfu.edu/mailman/listinfo/gretl-users

John C Frain
Economics Department
Trinity College Dublin
Dublin 2
mailto:frainj at tcd.ie
mailto:frainj at gmail.com

More information about the Gretl-users mailing list