Rajesh
You have to define an outcome variable and the variables that are the
characteristics.
In the following example:
psmatch2 trade carac1 carac2, out(outvar)
trade is the variable that define the treatment group,
carac1 and carac2 are the variables that define the characteristics of the
individuals,
outvar are the outcome variable.
The matching are made using carac1 and carac2 through the groups (training =
0 and training = 1).
The program estimate a probit model, calculate a propensity score and make
the matching based in this propensity. Trade is the dependent variable in
the probit model. Carac1 e carac2 are the covariates in this model. You have
the logit option, too. If you select the Mahalanobis option you don't need
to define the carac1 and carac2 variables as covariates and the sintax is:
psmatch2 trade, mahalanobis(carac1 carac2) out(outvar)
I think that nearest neighbor is the default matching criteria.
I hope to help with this.
Henrique
-----Mensagem original-----
De: [email protected]
[mailto:[email protected]] Em nome de Rajesh Tharyan
Enviada em: domingo, 16 de dezembro de 2007 11:22
Para: [email protected]
Assunto: st: psmatch2
Hi all,
I have a file set up as follows
coname mcap btom trade
A 35 .6 1
B 30 .1 0
C 32 .4 0
D 45 .3 0
E 56 .45 0
F 75 .23 0
G 26 .43 0
H 47 .35 0
I 25 .9 0
How can I use the propensity score matching technique to match the firms
for which trade=0 with firm for which trade=1, in terms of mcap and
btom?
Thanks
rajesh
*
* 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/
__________ Informagco do NOD32 IMON 2718 (20071212) __________
Esta mensagem foi verificada pelo NOD32 sistema antivmrus
http://www.eset.com.br
*
* 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/
