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]

st: IPTW with multiple categorical treatments


From   Ahu Alanya <ahu.alanya@student.hubrussel.be>
To   statalist@hsphsun2.harvard.edu
Subject   st: IPTW with multiple categorical treatments
Date   Fri, 24 Dec 2010 12:47:12 +0100

Dear all,

I am using Stata 10.1 and trying to solve the following problem.

In a social survey, we have there types of samples: Cooperatives ( C),
Reluctants (R ), and Non-respondents (N).  We have data on a number of
covariates for each sample. We want to weight cooperatives with
propensity scores to make the distribution of covariates similar to
that of the combined sample (C+R+N).

I ran a multinomial logit to estimate propensity scores for each
category (C,R,N). Then, used “predict” command for outcome=0 which is
being found in the Cooperatives sample; and finally used the inverse
of the estimated probability (1/p) as weights to weight Cooperatives.
Is that a correct approach to adjust covariate distribution in
cooperatives according to hypothetical combined sample?

(nr: 0=Cooperative 1=Reluctants 2=Non-respondents)

Stata code:

xi: mlogit nr i.agec i.educ i.soc poli
predict p, outcome(0)
gen inversp= 1/p if nr==0

Thanks a lot and happy holidays!

Ahu

*
*   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