stata - How to interpret results of teffects psmatch in State?

Question

I am having trouble figuring out whether psmatch is failing to match some observations and dropping them. I set a caliper of .2 and did not specify any other optional settings. I thought that Stata was supposed to throw an error if any observations were unmatched, but I am getting impossible min and max numbers of matches.

For instance, in a dataset with about 3x as many control observations as treatment observations, my output is:

Treatment-effects estimation                   Number of obs      =      3,093
Estimator      : propensity-score matching     Matches: requested =          1
Outcome model  : matching                                     min =         24
Treatment model: logit                                        max =        225

In another, with about 5x as many control observations as treatment observations I am getting:

Treatment-effects estimation                   Number of obs      =      2,558
Estimator      : propensity-score matching     Matches: requested =          1
Outcome model  : matching                                     min =          1
Treatment model: logit                                        max =          2

Neither of these is notifying me about unmatched observations. Does anyone know what might be going on here? Alternatively, is the number of observations reported in the output pre-match observations, and is there somewhere else I can find the post-match?

question from:https://stackoverflow.com/questions/65837089/how-to-interpret-results-of-teffects-psmatch-in-state

Answer 1

Waitting for answers