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]
[no subject]
Hi all,
I have used a multivariate regression model (with a number of continuous,
fixed and interacted variables) to produce a predicted value of my dependent
variable, which I will use as a propensity score in a matching exercise.
I want to loop through each "treated" case, and select a set of k-nearest
neighbors based on this (predicted value) "propensity score".
Ideally I would be able to save the id of each matched control case so that
I could use that in an analysis I will do later.
Although there are a number of propensity score matching algorithms
available (e.g., psmatch2, pscore, nnmatch - all available via ssc) I have
not found one that allows me to save an indicator variable (e.g., unique id)
for the matches (moreover their respective weights, in the case of identical
matches). Ideally I would have an output like geonear (available via ssc)
that allows me to save off an id for the k-nearest neighbors.
Is what I am hoping for available?
Ben
Ben Hoen
Staff Research Associate
Lawrence Berkeley National Laboratory
Office: 845-758-1896
Cell: 718-812-7589
[email protected]
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/