rdrobust: An R Package for Robust Nonparametric Inference in Regression-Discontinuity Designs

This article gives an introduction to the R package rdrobust. This package includes three main functions to conduct robust data-driven statistical inference in regressiondiscontinuity (RD) designs. The first and main function, rdrobust, implements conventional nonparametric RD treatment-effect point estimators and confidence intervals, as well as the robust bias-corrected confidence intervals proposed in Calonico, Cattaneo, and Titiunik (2013c) for local average treatment effects. This function covers sharp RD, sharp kink RD, fuzzy RD and fuzzy kink RD designs, among other possibilities. The second function, rdbwselect, implements several bandwidth selectors proposed in the RD literature. Finally, the third function, rdbinselect, implements a novel data-driven optimal choice of evenly-spaced bins, employs the resulting optimal bins to approximate the underlying regression functions by local sample means, and constructs the familiar RD plots usually found in empirical applications. A companion Stata package is described in Calonico, Cattaneo, and Titiunik (2013b). Preliminary and incomplete, comments welcom