I have a panel data and want to examine the relation between the passage of legislation (LawDummy, independent var) and debt usage (Debt, dependent var) using both the propensity score matched and full samples.

For the full sample, I run this regression: reghdfe Debt LawDummy Control, absorb(sic2) cluster(gvkey)

I want to construct a propensity score matched sample to mitigate the concern that treatment group differ inherently from control group. Each treatment group is matched (without replacement) to a counterfactual within the same two-digit SIC industry-year. Matched pairs are formed by identifying the pairings that result in observations with the smallest propensity score differences using some firm characters, ControlVar.

Anyone know how should I do this propensity score matching and use the matched sample to test my hypothesis?