Re: st: Heckman multinomial logit model

[Partha Deb <partha.deb@hunter.cuny.edu>]

If you are thinking of it as a treatment effects model, as distinct from 
a selection setup in which the outcome equation is not observed for one 
or more of the multinomial choices, you can also use -mtreatreg- .  
-findit mtreatreg- .

cheers,

Partha


Elisabeth Müller wrote:
> I would like to do an extension of Heckman.
>
> First, I have a multinomial logit model of the form:
> Yi=Zi g+ui
> where Yi= 0,1 or 2 (selection equation).
>
> Second, I have a general equation of the form
> Wj=Xj b+vj (regression equation).
>
> 1) How can I get maximum likelihood estimates from the selection equation?
> 2) Howcan I create Mills ratio from these estimates?
> 3) How can I do the regression equation with the Mills ratio?
>
> Thanks
>   

--
Partha Deb
Professor of Economics
Hunter College
ph:  (212) 772-5435
fax: (212) 772-5398
http://urban.hunter.cuny.edu/~deb/

Emancipate yourselves from mental slavery
None but ourselves can free our minds.
	- Bob Marley


*
*   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/