Package: cdid 1.2.0.1

cdid: The Chained Difference-in-Differences

Estimates group-time average treatment effects in panels with missing outcomes by the chained difference-in-differences of Bellégo, Benatia and Dortet-Bernadet (2025) <doi:10.1016/j.jeconom.2024.105783>. It is a layer on the 'did' package that keeps the att_gt() and aggte() interface and adds an estimation mode. Chained estimation, the default for panel data, combines short within-unit comparisons by the generalized method of moments. It provides chained inverse probability weighting (IPW), outcome regression (OR) and doubly robust (DR) estimation, where OR and DR use separate or pooled regressions of the controls' outcome changes; custom IPW callbacks; correction for covariate-dependent missing outcomes by observation weighting, change imputation or their augmented combination; parallel trends from each cohort's reference period (PT-post) or from a chosen common start date (PT-all); and a chained conditional pretrend test through conditional_did_pretest(). Ordinary mode delegates unchanged to the Callaway and Sant'Anna (2021) <doi:10.1016/j.jeconom.2020.12.001> estimators in 'did'.

Authors:David Benatia [cre, aut], Christophe Bellégo [aut], Joel Cuerrier [aut], Vincent Dortet-Bernadet [aut]