I am following Stata package to run the did_imputation of Borusyak,2021 in using the DiD imputation estimator

The normal DiD estimator in their regression is

did_imputation Y i t Ei [if] [in] [estimation weights] [, options]

where Y outcome variable

i variable for unique unit id

t variable for calendar period

Ei variable for unit-specific date of treatment (missing = never-treated)

In triple diffs, they define that

22) Triple-diffs: When observations are defined by i,g,t when, say, i are counties and g are age groups, specify a variable ig identifying the (i,g) pairs as the unit identifier, add appropriate FEs, and choose your clustering level, e.g.: . did_imputation Y ig t Eig, fe(ig i#t g#t) cluster(i) ... Note that the event time Eig should be specific to the i,g pairs, not to the i. For instance, Eig is missing for a never-treated age group in a county where other groups are treated at some point.

I am wondering the difference here is the difference of one subsample from another subsample or the difference of one subsample from the whole sample? Saying for an example, in my research, i is firm and g is developed and developing countries. So when using this approach, the tau retrieved from the regression is the difference between developed and developing countries or the difference between developed countries compared to the whole sample.

I guess the retrieving result mentioned the difference between developed and developing subsamples, but I am not totally sure so I am asking here.

Thanks in advance.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.