Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.

Re: st: mprobit with exclusion restriction

 From saqlain raza To "statalist@hsphsun2.harvard.edu" Subject Re: st: mprobit with exclusion restriction Date Wed, 23 Nov 2011 03:18:39 -0800 (PST)

```I studied the literature of Train (2003) about probit models. He says that

"With probit models, however, normalization for scale and level does not occur automatically. The researcher must normalize the model directly."

My question is How we can normalize the model? And how can we confirm that the error are really uncorrelated?

Thanks
Saqlain RAZA
PhD Student

From: Maarten Buis <maartenlbuis@gmail.com>
To: statalist@hsphsun2.harvard.edu
Sent: Tuesday, November 22, 2011 1:58 PM
Subject: Re: st: mprobit with exclusion restriction

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/  ;

*
*   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/
```