Re: st: iv analysis with first stage probit

["Mark Schaffer" <M.E.Schaffer@hw.ac.uk>]

J.B.,

From:           	<J.B.Durham@frb.gov>
Subject:        	Re: st: iv analysis with first stage probit
To:             	statalist@hsphsun2.harvard.edu
Copies to:      	Carl.Elliott@UCHSC.edu, Mark Schaffer <M.E.Schaffer@hw.ac.uk>,
       	owner-statalist@hsphsun2.harvard.edu, statalist@hsphsun2.harvard.edu
Date sent:      	Tue, 5 Nov 2002 11:10:50 -0500

> 
> Is this also applicable to 3 variable IV system?

No.  -treatreg- can handle a 2 equation system in which one equation 
has a single LHS endogenous dichotomous regressor, and the second 
equation is the probit for that dichotomous variable.

>  Do any iv commands
> have options for first stage probits?  Thanks.

Not that I'm aware of.

--Mark

> 
> 
>                     Mark Schaffer                                  
>                     <M.E.Schaffer@hw.ac.uk>           To:    
>                     statalist@hsphsun2.harvard.edu,
>                     Carl.Elliott@UCHSC.edu             Sent by:       
>                                       cc:     Mark Schaffer
>                     <M.E.Schaffer@hw.ac.uk>                           
>                       owner-statalist@hsphsun2.h        Subject:    
>                     Re: st: iv analysis with first stage probit       
>                                arvard.edu                             
>                          
> 
> 
>                     05/19/02 07:09 PM                                 
>                      Please respond to                                
>                       statalist                                   
> 
> 
> 
> 
> 
> 
> Looks equivalent to a treatment effects model.  Try -treatreg-.
> Should do what you want (and then some).
> 
> --Mark
> 
> Quoting Carl.Elliott@UCHSC.edu:
> 
> > I have run the following:
> >
> > ivreg wkcstchg p00age2 race educ gender income comor occupat
> > cidicat
> > (hp1m1gp4 = p00age2 race educ gender income comor occupat
> > cidicat interven),
> > first
> >
> > The output looks reasonable; however, the first stage
> > regression is a linear
> > regression and since the hp1m1gp4 variable is dichotomous I
> > would prefer the
> > first stage regression be a probit. I assume this can be done
> > by running a
> > probit and saving the predicted values to use in a linear
> > regression using
> > regress.
> >
> > Is there another way to do this that is less cumbersome?
> >
> > Thanks for your help.
> >
> > Carl
> >
> > *
> > *   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/
> >
> 
> Prof. Mark Schaffer
> Director, CERT
> Department of Economics, School of Management
> Heriot-Watt University, Edinburgh EH14 4AS
> tel +44-131-451-3494 / fax +44-131-451-3008
> email: m.e.schaffer@hw.ac.uk
> web: http://www.som.hw.ac.uk/ecomes
> ________________________________________________________________

Prof. Mark E. Schaffer
Director
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS  UK
44-131-451-3494 direct
44-131-451-3008 fax
44-131-451-3485 CERT administrator
http://www.som.hw.ac.uk/cert
*
*   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/