st: Re: Econometrically sound to use Mills ratio in a OLS estimation after mprobit?

[Luiz Gama <lcdgama@gmail.com>]

Dear Yuval,

Thanks for the tips.
My dependent variable is whether an individual is non migrant, return
migrant or non return migrant. Selmlog is only possible if only one
alternative has non missing values, so I can't use that.
I will think about the second alternative that you give me.
Thanks again,

-- 
Luiz Carlos Day Gama
Mestrando em Economia pelo Cedeplar/UFMG
*
*   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/