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: question simultaneous probit equations

 From "David Roodman (droodman@cgdev.org)" To "statalist@hsphsun2.harvard.edu" Subject RE: st: question simultaneous probit equations Date Fri, 30 Nov 2012 15:23:18 +0000

```cmp can't directly estimate simultaneous equations like that (yet). More generally, I'm not sure this model can be fully estimated anyway. In simultaneous equations you still need excluded variables in general, and if you have those then you can perform IV (perhaps with cmp). You could do a probit of Uh on Uw, instrumenting Uw with Y. Based on the model you show, I'm not sure you can go the other way.

You might also see the section of my paper about cmp that discusses logical consistency in multistage probit models. You can write down models like this that look OK but actually logically impossible.
--David

------------------------------

Date: Thu, 29 Nov 2012 20:38:07 +0100
From: Raluca Prelipceanu <raluca.prelipceanu@gmail.com>
Subject: st: question simultaneous probit equations

I am trying to estimate a model in which the wife's decision depends
on her husband's decision and vice versa (the husband's decision on
his wife's). They can either both decide to move or only the wife may
move or only the husband may move or both do not move.

utility of wife Uw (X, Y, Uh(X)) and utility of husband: U(X, Uw(X,Y)).

Can it be estimated by bivariate probit or should I use another technique?

And I have another question can I estimate a model in which I have a
probit in the first stage and then for the second stage a multivariate
probit? Could I estimate it using cmp?

Thank you,

Raluca

*
*   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/
```