Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Ben Hoen" <bhoen@lbl.gov> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | st: saving case ids when using "propensity score" matching |
Date | Tue, 20 Aug 2013 12:23:03 -0400 |
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 bhoen@lbl.gov http://emp.lbl.gov/staff/ben-hoen Visit our publications at: http://emp.lbl.gov/reports/re Sign up for our email list to receive publication notifications at: https://spreadsheets.google.com/a/lbl.gov/spreadsheet/viewform?formkey=dGlFS 1U1NFlUNzQ1TlBHSzY2VGZuN1E6MQ * * 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/