This discussion provides an excellent summary of TWFE event study model. However i have one estimation related question regarding dynamic TWFE model. In STATA we execute the the following code to obtain results on event study leads and lags:reghdfe Y F*event L*event, a(i t) cluster(i) where (F) and (L) are event leads and lags and (i) and (t) are unit and time fixed effects. This code gives me estimates on event leads and lags, but does not give me an overall causal parameter (overall ATT), which is a single coefficient obtained as an average of all the post-treatment lags. I want to ask, what i need to specify in the code above so that i get the coefficients on all event leads and lags (as before) but also get an overall ATT (single parameter- overall ATT)
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
1
Why not try xtevent with the diffavg option?
-
1$\begingroup$ Thanks, this seems a useful resource to read. $\endgroup$ Commented Sep 3, 2022 at 10:27