Re: st: question on gllamm for discrete latent variable with one factor structure

[Hey Sky <heyskywalker@yahoo.com>]

Dear Kolenikov

thanks for your answer and maybe for future answer too :) 

I think I have tried without the cons in the eq option with more than one latent variables .  
ie. 

eq mu1: edu
eq mu2: wrknt
eq mu3: wrkwt
eq mu4: home

gllamm dep indep,  i(id) base(4) link(mlogit) family(binom) nip(2) ip(f) nrf(3)  ///
 eq(mu1 mu2 mu3) dots trace

but it does not converge either. I will go back and check my memory is right or not. 

I will try the nocons option and see what the result will be. and thanks for remind me
the typo in my last post. I do have the name of the eq option in my code.

PS. what is the option peqs, geqs for? I think I do not really understand the part on manual

Nan
from Montreal



*
*   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/