Automated Local Regression Discontinuity Design Discovery

Authors:
William Herlands Carnegie Mellon University
Edward McFowland Iii University of Minnesota
Andrew Wilson Cornell University
Daniel Neill Carnegie Mellon University

Introduction:

This paper studies the problem of Inferring causal relationships in observational data. The authors develop the first statistical machine learning approach for automatically discovering regression discontinuity designs (RDDs), a quasi-experimental setup often used in econometrics

Abstract:

Inferring causal relationships in observational data is crucial for understanding scientific and social processes. We develop the first statistical machine learning approach for automatically discovering regression discontinuity designs (RDDs), a quasi-experimental setup often used in econometrics. Our method identifies interpretable, localized RDDs in arbitrary dimensional data and can seamlessly compute treatment effects without expert supervision. By applying the technique to a variety of synthetic and real datasets, we demonstrate robust performance under adverse conditions including unobserved variables, substantial noise, and model.

You may want to know: