Imputation Estimators

Imputation estimators of difference-in-differences uses not-treated/not-yet-treated observations to estimate Y(0). Then treatment effects are estimated using Y(1) - Yhat(0) (how aggregate treatment effects are estimated depends on the function).

did2s()

Calculate two-stage difference-in-differences following Gardner (2021)

Event Study 'Robustness Checks'

Uses the estimation procedures recommended from Borusyak, Jaravell, Spiess (2021); Callaway and Sant’Anna (2020); Gardner (2021); Roth and Sant’Anna (2021); and Sun and Abraham (2020)

event_study()

Estimate event-study coefficients using TWFE and 5 proposed improvements.

plot_event_study()

Plot results of event_study()

Example Data and DGP Function

gen_data()

Generate TWFE data

df_het

Simulated data with two treatment groups and heterogenous effects Generated using the following call: did2s::gen_data(panel = c(1990, 2020), g1 = 2000, g2 = 2010, g3 = 0, te1 = 2, te2 = 1, te3 = 0, te_m1 = 0.05, te_m2 = 0.15, te_m3 = 0)

df_hom

Simulated data with two treatment groups and homogenous effects Generated using the following call: did2s::gen_data(panel = c(1990, 2020), g1 = 2000, g2 = 2010, g3 = 0, te1 = 2, te2 = 2, te3 = 0, te_m1 = 0, te_m2 = 0, te_m3 = 0)

castle

Data from Cheng and Hoekstra (2013)