Causal Inference

Learning Goals

• Students can identify research questions that can be answered with applied econometrics and match them with possible research designs.
• Students understand the differences between correlation and causality and can identify possible threats to causality in empirical economic analysis.
• Students can implement a contemporary method of causal inference and critically evaluate and interpret the results.

Module Description

How do economists discern between correlation and causation in empirical analysis? This course offers training in research design and in selected methods used for causal inference in empirical economic research. It builds on the introductory courses in statistics and econometrics and focuses on formulating research questions, designing identification strategies, implementing statistical methods, and critically assessing empirical results. The emphasis is on the intuition of the different methods, their implementation, and the interpretation of the results. The main part of the course covers five methods from the toolkit of modern applied econometrics: matching, instrumental variables, regression discontinuity, difference in differences, and synthetic controls. The lecture is accompanied by a biweekly tutorial in which the methods are practiced in R. The course culminates in a short research proposal that outlines a feasible empirical analysis using one of the methods of the course and discusses its limitations.