Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Mlogit with panel


From   Michael Ingre <Michael.Ingre@ipm.ki.se>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Mlogit with panel
Date   Thu, 8 Jul 2004 09:53:21 +0200

On 2004-07-07, at 21.43, Clive Nicholas wrote:

Edlira Nazarani wrote:

I'm trying to run a mlogit using cross-sectional time series data.
Is it right to run as:
mlogit depvar indepvar , cluster i(ind)
or
should i use instead a xtlogit as:

xtlogit depvar indepvar,re i(ind)
If your dependent variable really is multinomial, my personal preference
would be to use -mlogit- but, if possible, to use time dummies on the RHS
of the equation. This should work especially if you have lots of
observations for each cross-section.
You could also estimate it as a random intercept model with -gllamm-.

. gllamm depvar indepvar , i(ind) link(mlogit) fam(binom) adapt trace

However, it may take some time to compute.

Michael

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index