36-729 TBD

6 units, Carnegie Mellon University

Observational Causal Inference refers to the study of cause-and-effect relationships without randomized treatment assignment. This course will explore issues of statistical estimation and inference in a variety of popular causal inference designs. These include randomized experiments?as a conceptual foundation?instrumental variables, regression discontinuity, difference-in-difference and matching. We will discuss statistical techniques such as randomization-based inference, sensitivity analysis and propensity scores. Based on time constraints and class interest, we may discuss graph-based causal search algorithms, causal mediation analysis, or spillover effects.

Prerequisites: none listed

Offered: Spring 2027

Courseweave needs JavaScript to show sections, workloads and planning tools.