st: RE: Mlogit for Panel Data

["Jacobs, David" <jacobs.184@sociology.osu.edu>]

I'm afraid your choices using Stata are limited.  There is no official Stata MLOGIT estimator for panel data.  

One work around that people use is to cluster on their caseid, which at least partially corrects the standard errors for the overtime repetition in panel data.

Another non Stata possibility is the econometric program called LIMDEP, which I recall does have a panel version for MLOGIT.

Dave Jacobs

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Amy May
Sent: Thursday, May 23, 2013 9:10 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: Mlogit for Panel Data

Dear Statlist
I have looked in the statlist archive but cannot find a specific answer to my question.

I would like to run a multinomial logistic regression model where the dependent variable has 3 outcomes,
0 is the basecase, 1, 2 ,3 are the alternatives.  The DV is not ordered, the numbers merely represent an auditor switching related choice.  The IVs are firm related characteristics such as balance sheet items and size.

My dataset is a panel with over 600,000 observations over a ten year period.  I would like to run a mutinomial logit with my dataset but after many hours of searching and reading up on this I cannot find a way to do this in STATA.  I have read up on gllamm and reprob but am unsure whether these would be suitable for my model.

Any answers would be greatly appreciated as I am really stuck!

Thanks
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/