1 to 1 propensity score matching using teffects

[Tom ODonnell]

Is there any way to perform 1 to 1 propensity score matching using teffects psmatch? It seems as though the default is a 1 to many match, and I have not been able to determine the best way to perform a 1 to 1 match instead.

[James Meltzer]

Tom,

My understanding is that the default is to do 1:1 matching. However, if you have a large dataset, there often will be ties for the same propensity score. The default is to match with all the ties. SSCC has published online a nice summary:

https://www.ssc.wisc.edu/sscc/pubs/stata_psmatch.htm

Hope that helps,
Jimmy