Re: st: RE: problems with gllamm

[Alexander Staus <alexander.staus@uni-hohenheim.de>]

for a mixed logit model you can also try -mixlogit- in stata from
"Hole AR. 2007. Fitting mixed logit models by using maximum simulated 
likelihood.  The Stata Journal 7: 388-401."

Alex



Nick Cox
> -gllamm- couldn't fit your model and is advising you to think about what
> you are asking. The usual advice is to try a much simpler model and
> build up. 
>
> Nick 
> n.j.cox@durham.ac.uk 
>
> Josep Maria Raya
>
> I'm trying to estimate a mixed multinomial logit, but when I use this  
> command (gllamm) in this form:
>
> gllamm motivacioport sexe dedat1-dedat5  
> dprocedencia_new2-dprocedencia_new5 usuariplatja dambqui1 dambqui2  
> dtransport1-dtransport3, expand( id chosen m) i(proc)  link(mlogit)  
> family(binom) nrf(4) eqs(a2 a3 a4 a5) nip(4) trace
>
> I obtain this error message:
> "could not calculate numerical derivatives
> missing values encountered
> (error occurred in ML computation)
> (use trace option and check correctness of initial model)"
>
> Any solution or the problem is that there is not solution?
>
> *
> *   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/
>
>   

--
Dipl.-Volkswirt
Alexander Staus
Institute for Agricultural Policy and Markets (420B)
University of Hohenheim
D-70593 Stuttgart
Germany
Ph: +49 (0)711-459-24304
Fax: +49 (0)711-459-22601
Email: alexander.staus@uni-hohenheim.de
Homepage: http://www.uni-hohenheim.de/marktlehre

*
*   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/