----- Original Message -----
From: "Rudy Loo-Kung" <[email protected]>
To: <[email protected]>
Sent: Friday, January 10, 2003 4:46 PM
Subject: st: How to get specific coef. in a fixed effects panel logit regression
>
> Greetings,
> I estimated a fixed effects panel logit regression using xtlogit and i need
> to recover the individual-specific intercepts. Does xtlogit allow to do so?
> I'd really appreciate some help.
> Thanks.
>
> Rudy Loo-Kung A.
> Asistente del �rea de Macroeconom�a y Pol�tica Econ�mica.
> Centro de Investigaci�n de la Universidad del Pac�fico.
I do not believe this is possible. The individual effects cannot be estimated
for a fixed T. From Hsiao (1986) "Unfortunately, contrary to the
linear-regression case, the MLEs for ai and B are not independent of each other
for the discrete choice models. When T is fixed, the inconsistency of ai is
transmitted into the MLE for B. Hence, even if N tends to infinity, the MLE of
B remains inconsistent"
The away around this incidental parameter problem is to maximize a conditional
likelihood function that sweeps away the individual effects.
I hope this helps,
Scott
Hsiao, Cheng (1986) "Analysis of Panel Data"
*
* 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/
