What you proposed was the method originally proposed by James Heckman,
and that is implemented by adding the -twostep- option. I don't know if
that is exactly what you want, I only translated the procedure you
proposed in Stata.
-- Maarten
--- "Armelini, Guillermo" <[email protected]> wrote:
> Thanks Maarten
>
> but what about the conditional regression. As far as I know Heckman
> model does not consider that the conditional regression is censored,
> that's why I proposed doing the procedure separately.
>
> Best wishes
>
> Guillermo
>
> ________________________________
>
> De: [email protected] en nombre de Maarten buis
> Enviado el: lun 12/24/2007 6:24
> Para: [email protected]
> Asunto: Re: st: estimating the covariance matrix of probit two stage
> procedure
>
>
>
> --- "Armelini, Guillermo" <[email protected]> wrote:
> > I'm trying to estimate a recursive simultaneous equation using a
> > probit model in the selection stage and a conditional censored
> > regression in the second stage. Specifically, the model is the
> > following one:
> >
> > Z(i)=AV(i)+U(i) (equation 1)
> >
> > Z(i)=1 if Z(i)>0
> >
> > Z(i)=0 if Z(i)<=0
> >
> > where A is a vector of parametes and V is a set of covariates for
> > each (i)
> >
> > Then I calculate Lambda which is the inverse mills of the predicted
> > values of equation one for those observations in which the
> Dependent
> > Variable is equal to 1
> >
> > The second step is to calculate the tobin model:
> >
> > Y(i)=bX(i)+E(i) if Z(i)=1 (second equation)
> >
> > In this equation "bs" are the coefficients to be estimated and "X"
> is
> > a vector of covariate which includes Lambda (which is the inverse
> > mills of the predicted values of equation one in which the DV is =
> 1)
> >
> > My question is how can I correct the covariance and the variance of
> > the parameters since, Lee, Maddala and Trost (1980) suggest that
> > everytime a two stage model like the one I describe before is
> > estimated, a procedure to correct the variance and covariance of
> the
> > parameters should be run.
> >
> > Is anybody aware about how to do this in Stata?
>
> Yes:
> heckman Y X, select(V) twostep
> X should not include lambda (Stata does that for you)
>
> Best whishes (to you and everybody else on the list),
> Maarten
>
> -----------------------------------------
> Maarten L. Buis
> Department of Social Research Methodology
> Vrije Universiteit Amsterdam
> Boelelaan 1081
> 1081 HV Amsterdam
> The Netherlands
>
> visiting address:
> Buitenveldertselaan 3 (Metropolitan), room Z434
>
> +31 20 5986715
>
> http://home.fsw.vu.nl/m.buis/
> -----------------------------------------
>
>
> ___________________________________________________________
> Support the World Aids Awareness campaign this month with Yahoo! For
> Good http://uk.promotions.yahoo.com/forgood/
> *
> * 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/
>
>
>
>
> This message has been scanned for viruses by TRENDMICRO,
> an IESE technology affiliate company and global leader in antivirus
> and content security software.
>
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
___________________________________________________________
Support the World Aids Awareness campaign this month with Yahoo! For Good http://uk.promotions.yahoo.com/forgood/
*
* 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/
