Machine Learning in Econometrics. The package R will be used throughout to demonstrate the techniques. The first half of the course will provide a practical introduction to modern high-dimensional function fitting methods — a.k.a. machine learning (ML) methods — for efficient estimation and inference on the treatment effects and structural parameters in empirical economic models. Participants will use R to immediately internalize and use most of the techniques in their own academic and industry work. Most of the lectures will be accompanied by the R-code that can be used to reproduce the empirical examples in the lectures during the lectures. The second half would cover additional unsupervised learning and deep reinforcement learning algorithms.
Outline: Review of classical regression for prediction and causal inference; Causal inference in approximately sparse linear structural equations models; Understanding of the inference strategy via the double partialling out and adaptivity; ML methods for prediction (reduced form estimation and evaluation of ML methods using test samples); ML methods for causal parameters, double ML for causal parameters in treat effect models and non-linear econometric models. Causal Panel with Machine Learning. Deep Reinforcement Learning, a.k.a. Artificial Intelligence in Economics