Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: xi question


From   Christopher F Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: xi question
Date   Mon, 7 Jun 2004 11:26:17 +0100

On Jun 5, 2004, at 7:33 AM, Andrea wrote:

This may be a basic question, but what I (still!) find puzzling is that
using the commands:

(a) xi: regress depvar varlist indivdummies timedummies

and

(b) xi: xtreg depvar varlist i.indiv i.time

I get the same fixed effects (which may be somewhat logical) but also
the same estimated coefficients for varlist, which I don't find that
logical... I'd expected to have different effects (at least on
varlist), given that the means are subtracted.
It sounds very fishy to include i.indiv in a fixed effects regression model (if that is indeed what you are doing: xtreg, fe) as those effects are already being incorporated in the model (by demeaning). For the general equivalency of regression with dummies vs the fixed effects transformation, see any good econometrics text (e.g. Greene).

Kit

*
* 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