RE: st: How do I impose the constraint that rho is a given number (eg 0.2) using biprobit?

["Martin Weiss" <martin.weiss1@gmx.de>]

<>


"* figure out the name of the coefficient
mat li e(b)"


The new -coeflegend- option may also be quite useful for this task:

*************
webuse school, clear
biprobit private vote logptax loginc years, coefl
*************


HTH
Martin


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Stas Kolenikov
Sent: Samstag, 20. März 2010 21:37
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: How do I impose the constraint that rho is a given number
(eg 0.2) using biprobit?

According to the help file, -biprobit- supports -constraints()-, so
you should be able to do what you want with that option. So extending
the help file example, we have

webuse school
biprobit private vote logptax loginc years

* find out the numeric value at the tanh scale of rho
di 0.5*ln( (1+0.2)/(1-0.2) )

* figure out the name of the coefficient
mat li e(b)

* define the constraint
constr def 11 _b[athrho:_cons]=0.20273255

* run the command with the constraint; never mind the error messages
biprobit private vote logptax loginc years, constr( 11 )

BTW, dear Stata Corp, if there's parsing of constraints within
-biprobit-, then these annoying error messages can be avoided by
suppressing the irrelevant constraints. But that is likely to be
difficult, and hardly affects the final result.

On Sat, Mar 20, 2010 at 1:57 PM, Claudia Berg <cberg@gwmail.gwu.edu> wrote:
> Hi,
>
> I'm a newcomer to statalist and hoped you could help me.  I'm trying
> to run a biprobit model with the constraint that rho (the correlation
> between the errors of the two equations) is fixed at a given value,
> say 0.2.  I found instructions online how to set rho to zero using
> heckman (see http://www.stata.com/support/faqs/stat/constraints.html)
> but am unclear how to modify it to fix rho at another value.
>
> Thank you!
>
> Claudia
> *
> *   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/
>



-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   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/