Re: [R] Method for selection bias with multinomial treatment

From: Barker, Chris [SCIUS] <cbarker1_at_scius.jnj.com>
Date: Mon 19 Jun 2006 - 07:22:25 EST


I don't know of R code per se, but
Check papers in the econometric literature by Heckman.  

And another good recent reference: Stat. Med. 2005 Jul 30;24(14):2197-212.

On the use of discrete choice models for causal inference Rusty Tchernis1;?; ?, Marcela Horvitz-Lennon2; 3 and Sharon-Lise T. Normand2; 4  

-----Original Message-----
From: r-help-bounces@stat.math.ethz.ch
[mailto:r-help-bounces@stat.math.ethz.ch]On Behalf Of ronggui Sent: Sunday, June 18, 2006 7:23 AM
To: r-help@stat.math.ethz.ch
Subject: [R] Method for selection bias with multinomial treatment

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 MatchIt package.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 R function to do the task.So I hope the lister can give me some suggestions.

Thank you in advance!

--
ٹ
Department of Sociology
Fudan University

	[[alternative HTML version deleted]]


______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html

Received on Mon Jun 19 07:28:58 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Mon 19 Jun 2006 - 08:12:13 EST.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.