Re: st: xtlogit vs. logit with dummies

["David W. Harless" <dwharles@vcu.edu>]

Sarit,
Unlike the cases of OLS and Poisson regression, fixed effects logit is not equivalent to 
logit with dummy variables for the cross-sectional units.  Logit with dummy variables for 
the cross-sectional units produces inconsistent estimates except in the unusual case where 
the number of observations per group is large.  There is a short entry on this in the 
Stata manual description for -clogit- under the section "Fixed-effects logit" (p. 224 of 
Version 9 Reference A-J manual) which gives some references.
Dave Harless

Sarit Weisburd wrote:

> Hello,
>
> I am confused by the large estimated differences I see when using
> xtlogit fe by(hid)
> and logit with dummies for hid. Does anyone know why the results are
> so different? Which one is more correct?
>
>
> Thanks,
> Sarit
> *
> *   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/
*
*   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/