It doesn't make sense, timing-wise, that the treatment affects the outcome and the outcome affects the probability of treatment. Perhaps what you envision is that pre-treatment unemployment duration affects the prob. of treatment, and then the treatment potentially affects post-treatment unemployment duration. Thus, you need to split unemployment duration into 2 variables - pre and post-treatment - where the former enters the model for the propensity score and the latter is the outcome of interest.
Dann
****************************************************
Daniel L. Millimet, Associate Professor
Department of Economics
Box 0496
SMU
Dallas, TX USA
phone: 214.768.3269
fax: 214.768.1821
web: http://faculty.smu.edu/millimet
****************************************************
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Patricia Vera Rojas
Sent: Tuesday, August 09, 2005 1:24 AM
To: [email protected]
Subject: st: psmatch2
Hi,
I am using psmatch2 in order to evaluate the impact of a
job assistance program in the unemployment duration of
participants.
Can the outcome variable be also included both in the
logit/probit and in the regression using psmatch2?
(the outcome variable is unemployment duration and it also
determines the probability of participating in the program)
Thanks,
Patricia
______________________________________________
Renovamos el Correo Yahoo!
Nuevos servicios, m�s seguridad
http://correo.yahoo.es
*
* 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/
*
* 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/
