Causal Inference and Personalized Decision Making

---

Topics and Objectives

• Potential outcomes, average treatment effects, and randomized trials
• Causal identification, propensity score weighting, and outcome regression
• Doubly robust estimation, cross-fitting, and statistical learning
• Unobserved confounding, instrumental variables, and weak instruments
• Conditional average treatment effects and meta-learners
• Optimal decision rules, policy value, and outcome weighted learning
• Variance reduction and continuous-treatment extensions
• Dynamic treatment regimes, backward induction, and policy learning

---

Module Schedule

During this three-week module, we will cover the following potential lecture materials as time permits. The first two weeks will focus mainly on causal inference, conditional average treatment effects, and outcome weighted learning. You will complete a set of homework assignments, due by Thursday of the second week. The third week will be a Present and Challenge week.

• [Randomized Control Trial] [Source Code]
• [Propensity Score and Inverse Probability Weighting] [Source Code]
• [Doubly Robust Estimation] [Source Code]
• [Conditional Average Treatment Effect] [Source Code]
• [Outcome Weighted Learning] [Source Code]

---

Homework

• TBA

---

Presentation Session

For this module’s presentation topics, you should select a paper on causal inference, heterogeneous treatment effects, personalized decision making, or another closely related topic.

Here are some candidate papers:

• TBA

---