EVENTSTUDYWEIGHTS: Stata module to estimate the implied weights on the cohort-specific average treatment effects on the treated (CATTs) (event study specifications)

eventstudyweights estimate weights underlying two-way fixed effects regressions with relative time indicators, It is optimized for speed in large panel datasets thanks to hdfe. To estimate the dynamic effects of an absorbing treatment, researchers often use two-way fixed effects regressions that include leads and lags of the treatment (event study specification). Units are categorized into different cohorts based on their initial treatment timing. Sun and Abraham (2020) show the coefficients in this event study specification can be written as a linear combination of cohort-specific effects from both its own relative period and other relative periods. eventstudyweights is a Stata module that estimates these weights for any given event study specification. For each relative time indicator specified in rel_time_list, eventstudyweights estimates the weights underlying the linear combination of treatment effects in its associated coefficients using an auxiliary regression. It provides built-in options to control for fixed effects and covariates (see Controls). eventstudyweights exports these weights to a spreadsheet that can be analyzed separately. This spreadsheet also contains the cohort and relative time each weight corresponds to, with headers as specified in cohort() and rel_time().