You should sort the data randomly before calling psmatch2...
set seed 123456
gen u=uniform()
sort u
psmatch2 ...
since there are identical propensity scores, if you call psmatch2 several times with data sorted in different orders you will get different answers since the matched pairs will differ. As long as the data are sorted randomly each time, and the other assumptions required by matching apply, each estimate is a consistent estimate of the ATT, even though they differ. By setting the seed and sorting the data randomly, you ensure you can replicate your results later.
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: Wednesday, August 31, 2005 12:31 AM
To: [email protected]
Subject: st: psmatch2
Hi,
I have used "psmatch2" (Matching Method = neighbor) and
I've got:
"There are observations with identical propensity score
values.The sort order of the data could affect your
results.Make sure that the sort order is random before
calling psmatch2"
What can I do to be sure the ATT shown is correct?
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/
