Basically I have many treated units and I would like to estimate (by applying the Diff-in-Diff methodology) the effect of the treatment. In my particular case it's very likely to observe certain heterogeneity in the "effect of treatment" across units, and that's why I would like to estimate a regression using time invariant characteristics of the units as covariates.

I haven't been able to find any article/text-book where this particular method is applied which makes me doubt about its theoretical validity.

(of course that in order to make everything comparable I will include control variables, or instead pre-match similar units in order to have more reliable conclusions).

Any suggestion? Thanks in advance

  • $\begingroup$ Can you clarify which estimates you are considering using as dependent variable? Predicted/fitted values from the diff-in-diff model? And are you asking about estimating a second regression using results from the first regression (main Diff-in-Diff model) as input? Would the second model include treatment variables, or what would it look like? Would your question in that second model still be what the effect of treatment is, or something else? $\endgroup$ – AlexK May 14 at 22:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.