Causal Machine Learning for Economic Analysis

3 credits · Graduate

This course bridges machine learning (ML) techniques and causal inference methods to address primarily economic questions. Students will explore key ML concepts such as penalized regression, regularization, cross-validation, and hyperparameter tuning, along with tree-based methods, and study their implementation within the context of economic research. The course integrates causal inference strategies grounded in the Rubin Causal Framework, focusing on ML-based approaches for both selection on observables and unobservables. Emphasis is placed on estimating both conditional average treatment effects (CATE) and heterogeneous treatment effects (HTE) using advanced techniques like double/debiased ML, causal forests, and meta-learners. Practical skills are developed through R-based labs, equipping students to combine ML with standard econometric tools for robust causal analysis and informed decision-making in economic applications. RESTRICTIONS: Graduate students in programs in the Department of Economics or Instructor permission EXCLUSIONS: ECON 4450 FORMATS: Lecture | Lab