I have a dataset containing treatment and control firms. Each firm belongs to a technology class defined in patentclass and innovation output of each firm is defined in patents which is a count data.
I want to match the control firms with treatment firms based on two criteria:
1) both treatment and control firm should be in the same patentclass
2) Control firms should have the same innovation output level based on patents. But I don't want to limit the match based on rage, I want to match with the firm with the closest match of patents
I can think of the following code but not sure how to include that restriction of closest match in patent matching.:
cem patentclass(#0) patent, treatment(treatment)
Any help is appreciated.
thanks
Related Posts with Coarsened Exact Matching (CEM)
change in sign and significance of linear term after adding quadratic and cubic termsHi all, I need your comments on change in sign and significance of linear term after adding quadrat…
Combining Cox Proportional Hazards model and Difference-in-DifferencesDear All, My question concerns whether it is possible to combine a Cox proportional hazards model w…
Flexible Regression Specification for Panel DataDear Stata community, I am currently looking to estimate a somewhat complex regression model to ana…
A question from an absolute beginner: hospitalization dataHi everybody. I have a set of hospital discharge records (time window 2001-2016) with the main diagn…
Matching observations across years for same idHi Forum, I am struggling with something very silly but would be very helpful if someone could help…
Subscribe to:
Post Comments (Atom)
0 Response to Coarsened Exact Matching (CEM)
Post a Comment