this response isn't really correct. multinomial models have been derived for panel data, using both random effects and conditional fixed effects approaches. unfortunately, these haven't been programmed in stata as far as I know.
-----Original Message-----
From: [email protected]
[mailto:[email protected]]On Behalf Of Clive Nicholas
Sent: Thursday, December 15, 2005 6:37 PM
To: [email protected]
Subject: Re: st: mlogit for panel
Nirina M wrote:
> Is there a way to use the same as "mlogit" command for panel data?
No, because it doesn't make sense. Think about this for a moment. Your
response variable has, say, four unordered categories which you want to
explain the variance of, but you can't estimate these together, precisely
because they are _unordered_, and thus _unmeasurable_.
In order to make this work in a panel data context, you would have to run,
say, four seperate panel models on 0/1 basis (i.e., category 1 against the
rest in model 1; category 2 against the rest in model 2; etc.).
Both the -xtlogit- and -clogit- routines should do what you're looking
for, so consult -help- on these.
CLIVE NICHOLAS |t: 0(044)7903 397793
Politics |e: [email protected]
Newcastle University |http://www.ncl.ac.uk/geps
Whereever you go and whatever you do, just remember this. No matter how
many like you, admire you, love you or adore you, the number of people
turning up to your funeral will be largely determined by local weather
conditions.
*
* 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/
