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: Propensity Matching


From   Sophie Jensen <thorns.n.roses@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Propensity Matching
Date   Wed, 29 Sep 2010 16:52:22 +0100

Dear statalisters,

I want to match a two samples of data, using info on two vars white
and fsm. I want to do a DD comparing the response of va24 on
outwage_male. So im comparing two sets of different samples - a
treatment and control. A policy was introduced in 2002 to change
things for the treatment but the introduction of these changes were
phased in in the treatment group, so I just use the last observation
as the 'before' period.

I'm not sure how to run the propensity matching regression, but I've
initially run the regression

psmatch2  Academy fsm eth_white va24 if year>2006, n(4) and get:
Probit regression                                 Number of obs   =       3101
                                                 LR chi2(3)      =      36.29
                                                 Prob > chi2     =     0.0000
Log likelihood = -54.892386                       Pseudo R2       =     0.2484

------------------------------------------------------------------------------
    Academy |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        fsm |   3.391206    .786663     4.31   0.000     1.849374    4.933037
  eth_white |  -.0178474   .4752928    -0.04   0.970    -.9494042    .9137094
       va24 |  -.0138131   .0195762    -0.71   0.480    -.0521816    .0245555
      _cons |  -3.625756   .5380517    -6.74   0.000    -4.680318   -2.571194
------------------------------------------------------------------------------

keep if year>2006

So Academy is my treatment and fsm and eth_white are what I want to
match on... va24 is y the dependent variable that we're testing for
change and before 2006 is the period 'before.'
I'm not sure how to interpret these results and for some reason the
command pstest treatment fsm white  doesnt work - what is the
necessity of this?  would appreciate your help immensely!

Thanks

*
*   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