Bookmark and Share

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   Michele Mancini <mmancini.stata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: simultaneous bivariate probit with mutual endogeneity
Date   Fri, 25 Feb 2011 15:41:11 +0000

Thank you for the answer. By the way is it correct to estimate the
recursive bivariate probit
Y1=a*Y2+b1X1
Y2=b2X2
using -biprobit-? I don't know if -biprobit- uses Full Information ML,
which I think is the best way to estimate such a model.


2011/2/25 Austin Nichols <austinnichols@gmail.com>:
> Michele Mancini <mmancini.stata@gmail.com>:
> -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
> <mmancini.stata@gmail.com> 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/
>

*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index