Causal Machine Learning
In this course, students will learn about the chances and challenges of using machine learning techniques to identify causal effects. The course consists of three main parts. The first part provides an introduction to causality, the counterfactual framework, and specific classical methods for the identification of causal effects. The second part deals with basics in supervised machine learning and introduces various regularization methods as well as tree-based machine learning methods. The third part combines the first two parts by discussing how machine learning can contribute to the estimation and identification of causal effects. The lecture will be accompanied by a bi-weekly tutorial, in which students will learn how to implement the discussed methods in statistical software.
Structure
1. Causality
- The counterfactual framework
- Causal graphs (DAGs)
- Randomized controlled trials
- Selection-on-observables strategies
- Regression
- Propensity score matching
- Inverse probability weighting
2. (Supervised) Machine learning
- Introduction to machine learning
- Regularization methods
- Tree-based methods
- Decision Trees
- Random forests
3. Causal machine learning
- Homogenous effects
- Post-double-selection
- Double-Machine-Learning
- Heterogeneous effects
- Double-Machine-Learning
- Causal trees
- Causal forests
Recommended readings
Athey, S. (2019). The Impact of Machine Learning on Economics. pp. 507-552 in Ajay Agrawal, Joshua Gans, and Avi Goldfarb (editors), The economics of artificial intelligence. University of Chicago Press
Boehmke, B. & Greenwell, B. (2019). Hands-On Machine Learning with R. Chapman and Hall. https://bradleyboehmke.github.io/HOML/index.html
Cunningham, S. (2021). Causal inference: The mixtape. Yale University Press. https://mixtape.scunning.com/
Mullainathan, S., & Spiess, J. (2017). Machine learning: an applied econometric approach. Journal of Economic Perspectives, 31(2), 87-106