Applied Machine Learning and Causal Inference

Graduate / Professional

Approaches to causal inference using machine learning. Covers randomized experiments with and without noncompliance, observational studies with and without ignorable treatment assignment, instrumental variables, and regression discontinuity. Machine-learning methods include bagging, boosting, tree-based methods such as random forests, and neural networks. Assignments provide students with hands-on experience with the methods. Applications are drawn from a variety of fields including political science, economics, public health, and medicine. Programming is central to the course and is based on the R programming language. Prerequisites: the equivalent of at least two of the following courses: S&DS 530, S&DS 538, S&DS 541, and S&DS 542; and previous programming experience (e.g., R, MATLAB, Python, C++), R preferred. Strong knowledge of OLS is assumed. Th 4pm-5:55pm