Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Greg Madison <gregmn2012@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Bootstrapped standard error for exact matching in psmatch2 |
Date | Wed, 6 Mar 2013 21:27:28 -0600 |
Hello StataUsers I am currently using psmatch2 for my research where I need to conduct exact matching(match agricultural farms within a county based on the propensity score) to estimate "att". A nice trick to do this had appeared previously in Statalist several times. To mention a few- http://www.stata.com/statalist/archive/2010-09/msg00073.html http://statalist.1588530.n2.nabble.com/PSMATCH-with-2-conditions-td5502216.html My concern is regarding estimation of bootstrapped standard error that incorporates the propensity score estimation and the fact that matching is done within a cell/exact matching.. Currently, psmatch2 can estimate such standard error if match is not done with exact match. For example, codes in psmatch2 for such matchign is given as Bootstrap including propensity score estimation: bs "psmatch2 training age gender, out(wage)" "r(att)" Bootstrap excluding propensity score estimation: bootstrap r(att): psmatch2 training, pscore(pscore2) outcome(wage) neighbor(5) I could not find anything on bootstrapped standard error while incorporating the stage with pscore estimation for exact matching. I would appreciate if anyone could suggest me on this. Regards * * 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/