Re: st: multinomial logit using -gllamm-

["Mary E. Mackesy-Amiti" <mmamiti@uic.edu>]

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/