Regression Discontinuity Designs

The regression discontinuity (RD) design is one of the most widely used nonexperimental methods for causal inference and program evaluation. Over last two decades, statistical econometric RD analysis have expanded matured, there now a large number methodological results identification, estimation, inference, validation. We offer curated review this literature organized around popular frameworks interpretation designs: continuity framework local randomization framework. For each framework, we discuss three main topics: ( a) designs parameters, focusing on different types settings treatment effects interest; b) estimation presenting based polynomial experiments, as well refinements, extensions, alternatives; c) validation falsification, summarizing an array mostly empirical approaches to support validity in practice.