Skip to contents

A workbench around the 'synthdid' estimator of Arkhangelsky, Athey, Hirshberg, Imbens and Wager (2021) doi:10.1257/aer.20190159 , built for designs with a single treated unit. Adds trimmed placebo inference for when a heterogeneous donor pool inflates the placebo standard error, covariate residualisation on untreated cells, a bias-corrected synthetic control with permutation inference, and fit diagnostics that respect the level shift synthetic difference-in-differences absorbs.

Author

Maintainer: James Hodges jamesdavidhodges@gmail.com