ATT in did_multiplegt_dyn command

[Dayla Spassio]

Hi all,

I am currently using the command did_multiplegt_dyn, a Stata command designed to estimate event-study Difference-in-Differences (DID) estimators in settings with multiple groups and multiple time periods.

The output provides the following estimates:

• Event-study treatment effects
• Average cumulative (total) effect per treatment unit
• Placebo estimates, which compare the outcome evolution of switchers and their controls before the switchers’ treatment changes for the first time.

While I understand the interpretation of these quantities within the dynamic DID framework, I would like to better understand how to conceptualize and compute the Average Treatment Effect on the Treated (ATT) in a context where the treatment is not binary but multivalued and adoption is staggered across groups and time.

Thanks