Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Fwd: Heckman model Vs treatment regression model


From   SMAC <s.afcha@ub.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Fwd: Heckman model Vs treatment regression model
Date   Tue, 13 Apr 2010 20:37:47 +0200

Dear Statalisters,

I have many doubts about wich econometric strategy I should use. I am trying
to estimate the impact of a program on a group of firms. I know that participation in the program is not random, so there is a selection bias. On the other hand, the participation in this program is directly related with the outcome variable Y, and this originate an endogeneity problem I think, since the probability to participate in the program is influenced by the outcome variable Y, and the outcome variable affects the probability to participate in the program and to be selected.

I begin with a two steps Heckman selection model:

in the first stage I estimate a probit:

S= xB+e
where S =1 if the firm apply for the program and is selected
and 0 otherwise. X is a set of explanatory variables

and in the second stage I estimate an OLS regression:

Y= zA+u
where Y is the outcome variable (a contiuos variable)

My doubts are, how can i calculate the impact of participation in the program?
Is this possible using Heckman post estimation commands? How?

Is Heckman selection model a correct approach or I should use treatreg command? I don't understand properly the difference between heckman selection model and the treatment effect model.

Thanks in advance for your help!

Moises






----- Final del missatge reenviat -----



*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index