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]

RE: st: Propensity Score Matching


From   <theodroe@bellsouth.net>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Propensity Score Matching
Date   Wed, 29 Feb 2012 07:58:14 -0500

Marco- Thanks for the feedback.

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Marco Ventura
Sent: Wednesday, February 29, 2012 2:38 AM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Propensity Score Matching

Misspecification problems have nothing to do with bootstraping. 
Bootstraping is required when the sample is too small and you do not 
rely on the closed form for computing the SE of the ATT.
Moreover, as far as I can see from your output the ATT is not 
significant and it will not become significant by simply bootstraping.
Finally, beware when using psmatch2 because it does not check for the 
balancing to hold, that means that the balancing must be checked by the 
Ichino and friend's routine, pscore.
Regards, Marco


Il 27/02/2012 14:51, theodroe@bellsouth.net ha scritto:
> I am new to Stata and propensity score matching. I have 143, 110 control
and
> 5,862 treated in my matched samples. I ran pscore2 to get this.
>
> Once I ran pscore I ran a radius matching method with the following
command:
>
> . attr englleve sch_grp sex_1 rd07leve, comsup boot reps(100) dots logit
> radius(0.0001)
>
> The process has given me:
>
> ATT estimation with the Radius Matching method
>
> Analytical standard errors
>
>
>
> ---------------------------------------------------------
>
> n. treat.   n. contr.         ATT   Std. Err.           t
>
> ---------------------------------------------------------
>
>
>
>       5862      143110      -0.073       0.008      -8.673
>
>
>
> ---------------------------------------------------------
>
> Note: the numbers of treated and controls refer to actual
>
> matches within radius
>
>
>
> The process has now run for 36 hours (MP2 version of Stata 12.1 AMD 2
core,
> WinXP pro 32byte). Should I let it run longer or am I hopelessly lost in
> some computational problem? Or did I miss specify the model?
>
>
> Thanks for your help and advice
> Ted
>
> *
> *   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/
*
*   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/

*
*   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