I have about 45 treatment firms and I am trying to create a control sample. Using that control 

sample, we plan to collect additional information about the firm, which would take some time and 

effort. I was wondering if I could use different CEM specifications for each treatment firm to get the 

most accurate control firms. 

The issue is that for some treatment firms, I get a huge strata, on the other 

hand, for other groups I have two control firms. I was wondering if I could use a more stringent 

matching criteria for those treatment firms with large number of control firms and a less stringent 

matching criteria for those that have a low number of control firms. 

My concern is how would I run a regression later because the weights are derived from different 

matching specifications. Or does it matter? 

Thank you for your time and consideration.

Kind Regards,

Pureum Kim