Chapter 1 Preface

This bookdown has been created based on the tutorials of the course 14.388 Inference on Causal and Structural Parameters Using ML and AI in the Department of Economics at MIT taught by Professor Victor Chernozukhov. All the scripts were in R and we decided to translate them into Python, so students can manage both programing languages. Jannis Kueck and V. Chernozukhov have also published the original R Codes in Kaggle. In adition, we included tutorials on Heterogenous Treatment Effects Using Causal Trees and Causal Forest from Susan Athey’s Machine Learning and Causal Inference course. We aim to add more empirical examples were the ML and CI tools can be applied using both programming languages.

The main topics this book covers are:

Prediction/Inference with High Dimensional Linear Models
Prediction in Modern Nonlinear Regressions (Random Forest and Deep Neural Networks)
Randomized Control Trials
Causal DAGs
Double/debiased Machine Learning
Heterogeneous Treatment Effects using Causal Trees
Heterogeneous Treatment Effects using Causal Forest
Feature Engineering With Deep Learning for Causal and Predictive Inference