Hi, everyone,
I found in the reference book that the conditional fixed-effect
estimation of xtlogit uses conditional logit only with grouping the time
variable (i.e., clogit ..., group (time_varname)). This is verified in
the program of xtlogit. But in Cheng Hsiao's textbook Analysis of Panel
Data (second edition) pp.196, it says that the alternative set is not
just the alternative of each time, but a new set with T!/s!(T-s)!
alternatives, where T is the time length and s is the times the
individual chooses 1. Can anyone explain this for me? Many thanks!
Gewei
*
* 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/
