st: re: Case-control matching_seek for help

["Ariel Linden. DrPH" <ariel.linden@gmail.com>]

Wu,

I would consider -optmatch2- for this activity (a user written program by
Mark Lunt -findit optmatch2-)

You can match across individual variables and choose how many controls per
case you would like. In addition you have a choice of distance measures to
choose from.

All that said, if you have a lot of variables to match on (you did not
provide us with that information), it would be preferable to generate a
propensity score and perform matching on that single value. If you choose
this approach, you can then use other programs that will allow you to choose
how many controls you'd like. -psmatch2-  (-findit psmatch2-) will allow you
to "match with replacement" so that you can use the same controls
repeatedly.

There are pros and cons to the various approaches, but at least this should
point you in the right direction.

Ariel 

Date: Tue, 5 Apr 2011 08:49:50 +0200
From: "Jesper Lindhardsen" <JESLIN01@geh.regionh.dk>
Subject: st: SV: Case-control matching_seek for help

Hi Wu,

I think I would look into some of the programs written for this purpose AND
are also documented; for example 

- -CEM- 
- -vmatch-

(use -findit- to install)
Also, if matching are to be exact(ie sex birthyear ect), you could use a
"merging" strategy instead.

However, the problem with insufficient controls means that ypu would either
accept it or use controls more than once - I am not sure how to do this, but
you could make a 'fake' stset and then use the stata command -sttocc-

HTH,

Jesper

*
*   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/