Footnote:
Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments September 13, 2019 erstellt
Description:
In isotonic regression discontinuity designs, the average outcome and the treatment assignment probability are monotone in the running variable. We introduce novel nonparametric estimators for sharp and fuzzy designs based on the isotonic regression. The large sample distributions of introduced estimators are driven by scaled Brownian motions originating from zero and moving in opposite directions. Since these distributions are not pivotal, we also introduce a novel trimmed wild bootstrap procedure, which does not require additional nonparametric smoothing, typically needed in such settings, and show its consistency. We find in Monte Carlo experiments that shape restrictions can improve dramatically the finite-sample performance of unrestricted estimators. To illustrate the empirical applicability of our approach, we estimate the incumbency effect in the U.S. House elections following an influential study of (Lee, 2008)