Hello,
I am working on a Difference-in-Difference panel model with year and entity fixed effects studying banking regulation. I wanted to carry out propensity score matching as in the pre-treatment my control and treatment groups differ significantly. I would like to start the matching procedure by running a probit regression with some explanatory variables for the calculation of propensity scores and then match each treatment bank to its 3 closest neighbors. Also, I would like the matching to be executed with replacement.
I am very unsure about the best way of approaching this in stata. I think ideally, I would like to create a subsample with new summary statistics of the matched sample and then run the DiD-fixed effects regression on this sample. Namely, I would like to produce a table like the one below and then run the main regressions for which my code is:
I would greatly appreciate your input and any guidance! 
Best,
Ulla
I am working on a Difference-in-Difference panel model with year and entity fixed effects studying banking regulation. I wanted to carry out propensity score matching as in the pre-treatment my control and treatment groups differ significantly. I would like to start the matching procedure by running a probit regression with some explanatory variables for the calculation of propensity scores and then match each treatment bank to its 3 closest neighbors. Also, I would like the matching to be executed with replacement.
I am very unsure about the best way of approaching this in stata. I think ideally, I would like to create a subsample with new summary statistics of the matched sample and then run the DiD-fixed effects regression on this sample. Namely, I would like to produce a table like the one below and then run the main regressions for which my code is:
Code:
reghdfe loanslongR NSFR_country_postNSFR L.depositratio L.totalcapitalratio L.lntotalassets L.LLRratio L.CIR gdp cpi, absorb(year id1) vce(cluster id1)
Best,
Ulla
