This volume introduces economists to modern machine learning techniques for establishing causal relationships from data. The book bridges the gap between predictive performance and the handling of unstructured data, covering essential topics such as automatic variable selection in high-dimensional settings, estimation of treatment effect heterogeneity, and natural language processing. It also explores synthetic control methods and macroeconomic forecasting. The authors provide a thorough theoretical grounding in machine learning methods, demonstrating their application in econometrics through numerous economic examples, programs, and exercises.

The book is positioned as a resource for economists aiming to integrate advanced machine learning into their causal inference and data analysis workflows. Its structure, which includes theoretical underpinnings alongside practical applications and exercises, suggests it is designed for both learning and implementation. The inclusion of topics like NLP and synthetic control indicates a focus on contemporary and sophisticated econometric challenges. The book aims to equip readers with the tools to establish causal relationships from complex datasets.