st: Re: inicial values not feasible

["Rodrigo Alfaro" <ralfaro76@hotmail.com>]

Maybe there is problem with the variable x, y and z.

Take a look of the correlations of these variables:

biprobit (a = b c d) (t = y z) if x!=.
corr x y z if e(sample)

Rodrigo.

----- Original Message ----- 
From: "Avcharian, Alexis" <avcharian@ort.edu.uy>
To: "stata" <statalist@hsphsun2.harvard.edu>
Sent: Thursday, March 30, 2006 4:10 PM
Subject: st: inicial values not feasible



> I estimated a biprobit model and stata said:
>
>   inicial values not feasible
>
> I tried different estimations and realized that the "inicial values not 
> feasible" problem only occured when i include variable x in the 
> estimation.
> i.e.:
> biprobit (a = b c d)(t = x y z)     inicial values not feasible
>
> If I don't include x, stata does the regression.
> It even works when I include x in the first equation (a= b c d x)(t= y z)
> Since x is a very important variable I have to include it in (t= x y z), 
> so dropping it is not an option.
> Can anyone please give me some advice.
> Thanks in advance.
> Alexis
>
>
>
>
>
>
> *
> *   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/