Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: probit/linear vs. treatreg

From   "Mark Schaffer" <>
Subject   st: probit/linear vs. treatreg
Date   Fri, 8 Aug 2003 17:46:28 +0100


Can you describe the two-step estimation procedure you're using and 
what you're doing with -treatreg-?

-treatreg- is a system estimator for a 2-equation system in which 
there is an equation for an endogeneous 1/0 variable, and an equation 
for a continuous dep var in which the endog 1/0 variable appears on 
the RHS.

If you run -treatreg- with the -twostep- option, it first estimates 
the equation for the 1/0 variable using a probit, and then estimates 
the equation for the continuous dep var a la Heckman.  This sounds 
like what you're doing by hand, so I'm not sure why there should be a 
difference.  Perhaps you ran -treatreg- without the -twostep- option? 
This would mean -treatreg- is doing a maximum likelihood estimation 
of the system, and so the results would indeed be different from a 
two-step estimation.


Date sent:      	Fri, 8 Aug 2003 11:54:12 -0400 (EDT)
From:           	Ronald Rutherford <>
To:             	"''" <>
Subject:        	st: Date: Fri, 8 Aug 2003 10:54:10 -0500 
Send reply to:

> Can anyone tell me why I get significantly different results using a two
> step probit/linear model for endogenity corrections versus the results
> obtained using the same set of regressors in a treatreg model?
>  Ron Rutherford
> *
> *   For searches and help try:
> *
> *
> *

Prof. Mark E. Schaffer
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
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index