re: st: Propensity Score Matching command in Stata 13

["Ariel Linden, DrPH" <ariel.linden@gmail.com>]

There are some differences between the new -psmatch- command in v13 and
-psmatch2- (a user written command by Edwin Leuven and Barbara Sianesi -
findit psmach2) that may matter to the user. 

For example, -psmatch- does not allow the user to choose between matching
with or without replacement. The command ONLY uses matches with replacement.


Also, I am not sure if -psmatch-allows the user to choose between matching
on the propensity score or matching on Mahalanobis distance metric. It
doesn't state it in the help file...

There may be other differences as well, but I have not yet used the new
command.

Ariel

Date: Tue, 25 Jun 2013 14:51:44 -0500
From: "Weichle, Thomas" <Thomas.Weichle@va.gov>
Subject: st: Propensity Score Matching command in Stata 13

Hi Statalisters,
For those who have been implementing propensity score matching using the
user-written command -psmatch2- by Edwin Leuven and Barbara Sianesi for
some time, it should be comforting to know that it appears as though the
new Stata 13 command -teffects- (more specifically "-teffects psmatch-")
performs matching using -psmatch2-.  In the Acknowledgments section of
the Stata online manual (http://www.stata.com/manuals13/te.pdf):  "We
thank Edwin Leuven and Barbara Sianesi for their inspirational command
psmatch2, which computes several of the estimators in teffects and a few
more that we have not implemented.".

Although I am not yet working on Stata 13, I look forward to comparing
the use of -psmatch2- and the new Stata command -teffects psmatch-.
Other comments and discoveries are welcome.

Tom Weichle
Math Statistician
VA Information Resource Center (VIReC)
Edward Hines Jr. VA Hospital, Bldg 18, 202E
708-202-8387 ext. 24261
Thomas.Weichle@va.gov


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/