Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: pooled vs panel data


From   Kit Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: pooled vs panel data
Date   Sat, 22 Jul 2006 09:00:15 -0400

If you mean that you did

regress y x1 x2 x3
xtreg y x1 x2 x3, fe

and the F-test at the foot of xtreg said that there are significant differences between the intercepts for the units of the panel, then the pooled OLS regression involves placing a number of inappropriate constraints on the fixed effects regression. When you place binding constraints on a regression, the coefficients change. When those constraints are soundly rejected by the data, it is a bad idea.

Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html


On Jul 22, 2006, at 2:33 AM, Jeff wrote:


I ran a few MLE regressions and set them up first as pooled data and second
as panel data. The regression coefficients were almost identical, only a
variation of a few decimal points, and the standard errors were much lower
in the panel regression. I was expecting the coefficients to still be an
unbiased estimate in both regressions, but I expected that they should be
slightly different from each other. Is this unusual?
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index