Applied Causality

Spring 2017, Columbia University

David M. Blei

Day/Time: Wednesdays, 2:10PM - 4:00PM
Location: 302 Fayerweather

Piazza site

Course Description

We will study applied causality, especially as it relates to Bayesian modeling. Topics include probabilistic graphical models, potential outcomes, posterior predictive checks, and approximate posterior inference. Each student will embark on a semester-long project around applied causal inference.

Reading assignments

  1. Introduction and logistics
  2. Potential outcomes
  3. Causal graphs
  4. Causal graphs
  5. Causal graphs and estimation
  6. Bayesian inference, potential outcomes, and randomization
  7. Bayesian inference, potential outcomes, and observational data
  8. Double robustness
  9. Instrumental variables
  10. Counterfactuals
  11. Genetic association
  12. Special Guest: Andrew Gelman
  13. Causality and medicine