You should write a short do file where you keep one metro area at a
time, do psmatch2, save the outcomes and the generated counterfactual,
compute the overall ATT/ATE/ATU by hand as the mean difference once you
have looped over all metro area, and then finally bootstrap the standard
error(s).
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 Hui-shien
Tsao
Sent: Friday, April 07, 2006 11:41 AM
To: [email protected]
Subject: st: psmatch2 and matching within the metropolitan area
Hi all ---
I am using psmatch2 to conduct propensity score matching. My data
is a pool of cases from four metropolitan areas. I would like to match
the treated tracts to the untreated tracts within the same metropolitan
area. I have tried to create dummy variables to control for
metropolitan areas. However, some of the matches still pick tracts from
a different metropolitan area. Is there any way I can match within the
same metropolitan area? Thank you for your help in advance.
Hui-shien
*****
Here is my syntax:
gen random=uniform()
sort random
psmatch2 treat metro1 metro2 metro3 var1 var2 var3 var4, logit
neighbor(1) common trim(3)
A simplified data structure:
id metro treat Var1 Var2 Var3
1 1 1 1 4 4
2 1 0 2 1 3
3 1 0 3 1 1
4 1 0 3 5 2
5 1 1 2 3 5
6 2 0 1 1 3
7 2 0 1 2 8
8 2 0 2 5 6
9 2 0 3 3 3
10 2 1 3 7 1
*
* 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/
