STAT24630 Causal Inference Methods and Case Studies
Description
Many of the most important questions in science, policy, business, and everyday life are causal rather than merely descriptive. Does remote work make employees more productive? How does a new tax policy influence economic activity? Does regular exercise improve mental health, or are healthier people simply more likely to exercise?
This course introduces the fundamental ideas and tools of causal inference, the branch of statis- tics that helps us answer these types of questions. We will build a foundation in the potential outcomes framework and study core methods including randomized experiments, observational studies, sensitivity analysis, and instrumental variables. Along the way, we will emphasize compu- tation, applications, and real-world case studies, such as the California alphabet lottery, evaluations of job training programs, clinical trials, and educational experiments.
By the end of the course, students will be able to critically assess causal claims, design and analyze studies, and apply statistical methods to disentangle causation from correlation in real data.
Textbook:
Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction by Imbens, Guido W. and Rubin, Donald B. (2015).
A First Course in Causal Inference by Ding, Peng (2023).
Grading
- Homework assignments: 40%
- There will be 4 assignments in total.
- Late homework will not be accepted for grading (medical emergencies excepted with proof).
- Homework will be submitted through Gradescope and is due at 11:59pm the due date.
- 2 Quizzes: 15%
- Final group project + presentation: 35%
- Final self-assessment: 10%