That's what I thought but I wondered if there is a way to change the initial
values used by biprobit (it doesn't seem to be such an option in the command).
I guess from you answer that there is not.
Many thanks, Nestor
"Marc F. Bellemare" ha escrito:
> The problem is that the initial values that Stata plugs into your
> likelihood function are not feasible. My hunch is that you'll have to use
> the -ml- set of commands and program a bivariate probit yourself so as to
> be able to tweak the algorithm so that (i) it has feasible initial values;
> and (ii) converges.
>
> I hope this helps,
>
> Marc
>
> At 05:58 PM 11/25/2004 -0300, you wrote:
> >Dear statlisters,
> >I'm estimating several bivariate probit model using the biprobit
> >command.
> >In some of them I recieve the follwing error message:
> >
> >"initial values not feasible"
> >
> >Does this mean that I can't estimate those models?
> >Does anybody know how this can be solved?
> >
> >Thanks,
> > Nestor
> >
> >*
> >* For searches and help try:
> >* http://www.stata.com/support/faqs/res/findit.html
> >* http://www.stata.com/support/statalist/faq
> >* http://www.ats.ucla.edu/stat/stata/
>
> __________
> Marc F. Bellemare
> Ph.D. Candidate
> Department of Applied Economics and Management
> Cornell University
> Ithaca, NY 14853-7801
> http://www.people.cornell.edu/pages/mfb24
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
