Propensity score matching

[Nicola Rossi]

Dear Stata Users,

I am trying to create a new control sample based on the propensity score matching procedure.

In particular, I have a man sample (e.g. 100 id) and a control sample (e.g. 10.000).
I have followed this procedure.

1. Logit model. logit (dummy_MainSample_CSample) covariates
2. predict psscore
Now, I have an additional column with the psscore in my dataset.
I want to create a new dataset with 100 id (main sample) and 100 id from the control sample based on the psscore (this control sample should include the id with most similar characteristics to the main sample).

Any suggestions would be really appreciated,
Thank you in advance,
N

[Andrew Musau]

There are problems extracting a matched sample and doing analysis using this and the control group. From https://www.ssc.wisc.edu/sscc/pubs/stata_psmatch.htm

Sometimes researchers then want to run regressions on the "matched sample," defined as the observations in the treated group plus the observations in the control group which were matched to them. The problem with this approach is that the matched sample is based on propensity scores which are estimated, not known. Thus the matching scheme is an estimate as well. Running regressions after matching is essentially a two stage regression model, and the standard errors from the second stage must take the first stage into account, something standard regression commands do not do. This is an area of ongoing research.

However, the same link shows you how to extract the matched sample using psmatch2 from SSC.

[George Ford]

could try

cem psscore, tr(treated_dummy) k2k

reg y treated if cem_matched