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: analysis on PSM matched control group


From   "Sara Abrams" <[email protected]>
To   <[email protected]>
Subject   RE: st: analysis on PSM matched control group
Date   Thu, 5 Aug 2010 12:43:46 -0400

Thank you, Neha.  I don't think _treated is the variable I'm looking for
though.  

When I run the frequency of _treated for one of my PSM matches, I have
1,258 Untreated observations and 16 Treated observations.  I am looking
for a variable (or some combination of varaibles) that will identify
only those Untreated cases used as control matches to the Treatment
group.  In this case, I'm doing a 1:3 match, so I would expect about 48
cases, but probably less since there is some degree of replacement. 

-Sara



-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Neha Nanda
Sent: Monday, August 02, 2010 2:41 PM
To: [email protected]
Subject: Re: st: analysis on PSM matched control group

Hi Sara 

When you use psmatch2, several variables are created, one of them is
called _treated. You should see this variable through the data browser
once. It mentions treated or untreated for every treated or control
observation on or off the common support. 

I guess this should enable you to identify the control observations. 

Hope this helps. 

Neha Nanda 
Ph.D Candidate 
Florida State University 
Tallahassee
Fl-32304



----- Original Message -----
From: Sara Abrams <[email protected]>
Date: Monday, August 2, 2010 11:29 am
Subject: st: analysis on PSM matched control groupS
To: [email protected]

> I've conducted some PSM analysis using the psmatch2 procedure (1:3
> nearest neightbor matching, logit model).  I need to run 
> frequencies and
> crosstabs for covariates used in matching by my outcome variable for
> both the unmatched and matched control groups.  How do you identify 
> thematched controls in the dataset after PSM in order to conduct 
> analysiswith just this susbset of cases?  How do you handle matched 
> cases that
> have been matched with more that one treatment case (with 
> replacement)? 
> 
> Thanks,
> 
> -Sara
> 
> 
>       
> 
> 
> 
> 
> 
> *
> *   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index