STAT 286/GOV 2003: Causal Inference with Applications

 

  Syllabus

Substantive questions in empirical scientific and policy research are often causal. Does voter outreach increase turnout? Are job training programs effective? Can a universal health insurance program improve people's health? This class will introduce students to both statistical theory and applications of causal inference. As theoretical frameworks, we will discuss potential outcomes, causal graphs, randomization and model-based inference, sensitivity analysis, and partial identification. We will also cover various methodological tools including randomized experiments, regression discontinuity designs, matching, regression, instrumental variables, difference-in-differences, and dynamic causal models. The course will draw upon examples from political science, economics, education, public health, and other disciplines.

This course was listed as Stat186/Gov2002 in 2018 and 2019.

  Lecture Slides

Module 1: Potential Outcomes (slides, videos)
Module 2: Permutation Test (slides, videos)
Module 3: Average Treatment Effects (slides, (videos)
Module 4: Linear Regression and Randomized Experiments (slides, videos)
Module 5: Instrumental Variables (slides, videos)
Module 6: Regression Discontinuity Designs (slides, videos)
Module 7: Observational Studies (slides1, slides2, videos)
Module 8: Matching Methods Methods (slides, videos)
Module 9: Causal Mechanisms (slides, videos)
Module 10: Fixed Effects, Difference-in-Differences, and Synthetic Control Methods (slides1, slides2, videos)
Module 11: Heterogeneous Treatment Effects (slides, videos)

© Kosuke Imai
 Last modified: Sat Aug 20 17:30:54 EDT 2022