Re: st: Fixed effects vs. Random effects

["Susan M. Richter" <srichter@primal.ucdavis.edu>]

Shon

You might want to look at
Hamerle and Ronning, 1995 "Panel Analysis for Qualitative Variables" in 
the _Handbook of Statistical Modeling for the Social and Behavioral 
Sciences_.
They provide a summary on sample size issues with a fixed effects logit 
model. 
Furthermore since the xtlogit with fe is essentially a clogit you can 
look up the references cited in the Stata Reference Manual for clogit.   
They provide excellent background information on the disadvantages (and 
advantages) of using the xtlogit, fe and is where you will find the 
Hamerle and Ronning reference.

Susan

ILR School wrote:

> Does anyone know where I can find out the minimum
> number of panels one needs to run a fixed-effects
> model?  I have a dataset of 800 firms with only 3 time
> periods using a xtlogit regression model. I am
> currently using a random effects model.
>
> Thank you,
> Shon R.
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com 
> *
> *   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/