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 on April 23, and its replacement, statalist.org is already up and running.


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

st: Diff-in-Diff Question


From   "Scott Holupka" <scott.holupka@jhu.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Diff-in-Diff Question
Date   Fri, 26 Apr 2013 10:37:13 -0400

Fellow Statausers:

I have a question about analyzing some panel data using
difference-in-difference estimates.

Assume I have a panel dataset where:

Tr = treatment variable
Tm = time variable (0=baseline)
Covl = list of covariates
id = case id

I can compute difference-in-difference estimates using the model:

1. reg outcome tr tm c.tr#c.tm `covl'

I next want to estimate a difference-in-difference model using kernel
propensity score matching.  For this I first estimate the propensity model
using the user-written program PSMATCH2:

2. psmatch2 tr `covl' if panel==0, kernel

But I'm not sure how to use the output from (2), the weight variable, in my
difference-in-difference regression.

Please note that I've used the user-written program DIFF to compute (1) and
it gives me identical results.  DIFF also computes kernel matched
difference-in-difference scores.  However I want to add a 3-way interaction
to my models which DIFF does not handle which is why I cannot rely on just
the DIFF program.

Any help would be greatly appreciated.

Scott Holupka


*
*   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/


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