Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: RE: Propensity Score Matching


From   Fernando Rios Avila <f.rios.a@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: Propensity Score Matching
Date   Sat, 12 May 2012 11:13:05 -0400

Hi Carlos
So one thing you need to know is that there is no RULE to structure
the model for a propensity score matching setup. And unless one is
working with your data, its going to be impossible to just give you a
model that will solve your problem.
As you may know, propensity score matching (PSM) relies on finding
observations who are treated and untreated who are similar enough to
be used as counter factual. The error you reported just means that
based on -pscore- analysis, after estimating the propensity score and
matching, some characteristics are sill statistically different.
What it is suggested is to try a different specification, which
basically means you have to either include more variables as
covariates, or include either higher terms (square cubes) and or
interactions.
Again, there is no magical rule, you just need to keep trying until
one of the specification works for you.
Bests.
Fernando
On Sat, May 12, 2012 at 11:01 AM, CARLOS MAGNO SANTOS CLEMENTE
<carlosmagno_sc@ig.com.br> wrote:
> I am using the following command:
>
> "pscore binaria pop_2010 pib_09 s_inad_10 analf_10 medi_ind_
> decl_med_2  decl_med_2 dista_capi, pscore(mypscore) blockid(myblock)
> comsup level (0.005) logit"
>
>
> binaria = dummy
> pop_2010 pib_09 s_inad_10 analf_10 medi_ind_  decl_med_2  decl_med_2
> dista_capi = covariables
>
> Could help me in structuring the model?
>
> Thank you
>
>
>
>
> 2012/5/11 Sarah Edgington <sedging@ucla.edu>:
>> Carlos,
>> The glib answer is that the message you're getting already tells you what
>> procedure to take "Try a different specification of the propensity score."
>> As noted in the Stata FAQ
>> (http://www.stata.com/support/faqs/res/statalist.html#question) you should
>> specify what command you're actually using, and where it comes from if it
>> isn't an official Stata command.  Without that information and at least a
>> summary of what you're trying to do, it's going to be very difficult to
>> offer any more specific advice.
>>
>> Have you read the article "Estimation of average treatment effects based on
>> propensity scores" by Becker and Ichino, which appears in the Stata Journal
>> (2002 v2 #4)?  http://www.stata-journal.com/sjpdf.html?articlenum=st0026
>> I found that discussion helpful when implementing an analysis with
>> propensity scores.
>>
>> Good luck.
>>
>> -Sarah
>>
>> -----Original Message-----
>> From: owner-statalist@hsphsun2.harvard.edu
>> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of CARLOS MAGNO
>> SANTOS CLEMENTE
>> Sent: Friday, May 11, 2012 3:41 PM
>> To: statalist@hsphsun2.harvard.edu
>> Subject: st: Propensity Score Matching
>>
>> PSM the following error occurs.
>>
>> he balancing property is not satisfied
>>
>> Try a different specification of the propensity score
>>
>> What procedure to take?
>> --
>> Att,
>>
>> Carlos Magno Santos Clemente
>>
>> Geógrafo, Mestrando pela Universidade
>>
>> Estadual de Montes Claros – UNIMONTES
>>
>> Atuação: Geotecnologias
>>
>> *
>> *   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/
>
>
>
> --
> Att,
>
> Carlos Magno Santos Clemente
>
> Geógrafo, Mestrando pela Universidade
>
> Estadual de Montes Claros – UNIMONTES
>
> Atuação: Geotecnologias
>
> *
> *   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index