Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Methods for selection-bias when treatment is multinomial

From   "jean ries" <>
Subject   RE: st: Methods for selection-bias when treatment is multinomial
Date   Mon, 19 Jun 2006 14:03:23 +0200

As far as I can remember, the methods developed in the paper you cite
below are quite straightforward to implement using a software like
Stata. Rather than estimating the propensity scores via a binary logit
or probit, you should use their multinomial counterparts.  Then you
can use these propensity scores in routines like -psmacth2- or -att*-
( type -search pscore- to find the att* programs).

Here is another paper you should read if you are interested in these methods:

Hope this helps,


Date: Sun, 18 Jun 2006 01:43:46 +0800
From: ronggui <>
Subject: st: Methods for selection-bias when treatment is multinomial

I have to the treatment effect base on the observational data.And the
treatment variable is multinomial rather than  binary.Because the
treatment assignment is not random,so the selection-bias exists.Under
this condition,what's the best way to estimate the treatment effect?

I know that if the treatment is binary,I can use propensity score
matching using psmatch2 routine.But what about multinomial?

Maybe one way is the method  proposed by Imbens,2000,The role of the
propensity score in estimating dose-response
functions,Biometrika,87:706-710.But I can't find any Stata routine to
do the task.So I hope the lister can give me some suggestions.

*   For searches and help try:

© Copyright 1996–2020 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index