Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: simultaneous bivariate probit with mutual endogeneity
From
Austin Nichols <[email protected]>
To
[email protected]
Subject
Re: st: simultaneous bivariate probit with mutual endogeneity
Date
Fri, 25 Feb 2011 10:28:03 -0500
Michele Mancini <[email protected]>:
-biprobit- can't do that--nor can -cmp- (on SSC)--you will have to
program it yourself; see
http://www.stata.com/help.cgi?ml
http://www.stata.com/bookstore/ml4.html
On Fri, Feb 25, 2011 at 9:05 AM, Michele Mancini
<[email protected]> wrote:
> Hi Statalisters,
>
> I have to estimate the following bivariate probit:
>
> Y1=a*Y2+b1X1
> Y2=c*Y1+b2X2
>
> where X1 and X2 are two sets of different variables (some of them are
> the same) and Y1 and Y2 are two [0,1] endogenous variables.
> I don't have any problem using -biprobit- but there is a problem
> indeed: coherency conditions. The literature says that a coherency
> condition is a*c=0.
> With this condition you force the system to be recursive, setting for
> example c=0.
> Imposing c=0 and using -biprobit- is fine: high significance and rho
> is different from zero (p=0.0000).
> But I'd like to try another way. I've read this paper (Hajivassiliou
> 2008, pag.15: http://www.nuffield.ox.ac.uk/General/Seminars/Papers/969.pdf):
> imposing a>0 and c<0 you have solved the problem, using conditional
> MLE in a way that the observed LDVs to lay outside the empy region
> (see pag. 20-21 of the paper).
> How can I perform this using -biprobt- or other Stata commands?
*
* 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/