st: RE: RE: psmatch2-identifying matched pairs

["Garth Rauscher" <garthr@uic.edu>]

I managed to figure out how to identify and subset my data to include only
the matched pairs created by psmatch2. What was (and still is) confusing me
was that if I included the ate option, some of the matched pairs were
"broken" and a member of the pair was no longer identified as such. I don't
understand why this happened, but, nonetheless for my purposes, leaving off
the ate option allowed the following code to work: 

---------------------------------------------------------------------
psmatch2 path2, pscore(p_prodterms) outcome(binarystage) caliper(.001)
noreplace neighbor(1)

gen pair = _id if _treated==0 
replace pair = _n1 if _treated==1
bysort pair: egen paircount = count(pair)
drop if paircount !=2
---------------------------------------------------------------------

This code creates the variable pair that identifies the matched pairs.


I forgot to mention- Hat tip to the authors of psmatch2 for providing this
partial solution...

------------------------------------------------------------------------
The following code:

gen pair = _id if _treated==0 
replace pair = _n1 if _treated==1

creates a variable that identifies the matched pairs but also includes those
within the region of support who are not included in a matched pair. I
cannot figure out how to drop those observations that are not part of a
matched pair. Any help would be appreciated. 


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
Sent: Wednesday, March 31, 2010 8:13 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: RE: psmatch2-identifying matched pairs

I don't know anything about -psmatch2- (which is a user-written program
from SSC, as should be explained). But the problem may be soluble by
-egen, group()-. A more general discussion is given by 

SJ-7-4  dm0034  . . . Stata tip 52: Generating composite categorical
variables
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  N.
J. Cox
        Q4/07   SJ 7(4):582--583                                 (no
commands)
        tip on how to generate categorical variables using
        tostring and egen, group()

Nick 
n.j.cox@durham.ac.uk 

Garth Rauscher

I apologize if this has been answered before but I could not find the
solution in the archives. When I perform 1-1 matching using psmatch2,
several new variables are added to my dataset. Among these are _id,
which is
a unique identifier for each observation, and _n1, which identifies the
_id
for the observation in the matched pair. I'd like to be able to define a
new
variable that uniquely identifies each matched pair. In other words if
there
are 100 observations making 50 matched pairs, there are 100 unique
values of
_id and 100 unique values of _n1. I'd like a variable with 50 unique
values,
one for each matched pair. Stata doesn't appear to create this variable.
Having this variable would enable me to conduct stratified analyses
(stratifying on matched pairs) outside of what is provided by psmatch2.
Given _id and _n1 there is probably a way to do this but it is beyond my
skill set.  I'd appreciate if there are any ideas out there as to how to
go
about defining this variable.


*
*   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/


*
*   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/