Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

re: st: Manually compute IPTW for PS matching?


From   "Ariel Linden, DrPH" <ariel.linden@gmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   re: st: Manually compute IPTW for PS matching?
Date   Fri, 16 Mar 2012 10:56:05 -0400

Pete, what you are attempting to do doesn't make much sense. 

1:1 matching results in an "average treatment on the treated" (ATT)
estimator. You then want to take this matched sample - which has just been
"adjusted" - and then adjust it to an "average treatment effect" (ATE)
estimator? 

You'd use the inverse probability of treatment weight (IPTW) on the whole
population after estimating the propensity score on that population. The
IPTW will then give you an ATE estimate.

To calculate the IPTW:

gen ipt = cond(treatvar, 1/propvar, 1/(1-propvar))

Ariel 

Date: Fri, 16 Mar 2012 13:03:54 +0700
From: TA Stat <tastat@gmail.com>
Subject: st: Manually compute IPTW for PS matching?

Dear All

This is a simple question.  After run psmatch2 command (1:1 matching),
I got pscore, treated, support, weight, n1, nn, pdif variables.  Do I
have to manually calculate IPTW? Is any recommended step-by-step
illustration of PS matching in Stata available?

Thanks
Pete

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


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index