Re: st: mprobit with exclusion restriction

[Maarten Buis <maartenlbuis@gmail.com>]

On Tue, Nov 22, 2011 at 1:46 PM, saqlain raza wrote:
> I am using -mprobit- for my data that has 4 alternatives (1,1), (1,0), (0,1) and (0,0) and I want to see the results of models as
>
> model_1        alpha(0) + alpha(1)X(1) + alpha(2)X(2)
> model_2        beta(0) + beta(1)X(1) + beta(2)X(2) + beta(3)X(3) + ...... + beta(15)X(15)
> model_3        gamma(0) + gamma(1)X(1) gamma(2)X(2) + ........ + gamma(13)X(13)
>
>
> whereas the model_4 will be considered as reference model. How could it be possible? I know that with -mvprobit- it can be done but my interest is is multinomial probit model.

Notice that -mprobit- only fits a very special case of the multinomial
probit model: it constraints the correlations between the error terms
to be 0, so you do not need an exclusion restriction to estimate it.

Assuming that you still want to do this, I would add all explanatory
variables to your -mprobit- model and constrain the coefficients to be
0 when you do not want to use that variable in that equation. You can
see examples on how to do that in the -mlogit- case in -help
constraint-.

Hope this helps,
Maarten

>
>
> Saqlain RAZA
> PhD Student
>
> *
> *   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/
>



-- 
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany


http://www.maartenbuis.nl
--------------------------

*
*   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/