did

Assess how robust difference-in-differences results are to violations of parallel trends using the honestdid package in Stata, progressing from a simple 2x2 DiD to multi-period event studies with relative magnitudes and smoothness restrictions