Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: multinomial logit using -gllamm-


From   "Mary E. Mackesy-Amiti" <[email protected]>
To   [email protected]
Subject   Re: st: multinomial logit using -gllamm-
Date   Tue, 08 Mar 2011 16:38:52 -0600

My guess would be that the first model implies equal variance (random effects) across levels.


On 3/8/2011 10:23 AM, Jeph Herrin wrote:
The short answer is that I specified (but failed to say so here) the baseline category in both cases. The long answer is that in generating the variable -alt- I miscalculated so that the categories shifted one. So I had it wrong.

As I am still re-estimating the model, I don't know yet if that will resolve the first half of the problem. Even if so, I would still be curious to know
what is being estimated in the simpler model specification.

thanks,
Jeph


On 3/8/2011 11:16 AM, Brendan Halpin wrote:
Just a thought: is the baseline category implied by the "alt" strategy
the same as in the first model?

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



--
Mary Ellen Mackesy-Amiti, Ph.D.
Research Assistant Professor
Community Outreach Intervention Projects (COIP)
School of Public Health m/c 923
Division of Epidemiology and Biostatistics
University of Illinois at Chicago
ph. 312-355-4892
fax: 312-996-1450

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


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index