A Tutorial on Doubly Robust Learning for Causal Inference
It addresses the problem of complexity and inaccessible software for data scientists and statisticians, but it is incremental as it focuses on education and application rather than new research.
This tutorial tackles the limited practical adoption of doubly robust learning for causal inference by demystifying the methods and demonstrating their application using the EconML package, aiming to make them accessible to researchers and practitioners.
Doubly robust learning offers a robust framework for causal inference from observational data by integrating propensity score and outcome modeling. Despite its theoretical appeal, practical adoption remains limited due to perceived complexity and inaccessible software. This tutorial aims to demystify doubly robust methods and demonstrate their application using the EconML package. We provide an introduction to causal inference, discuss the principles of outcome modeling and propensity scores, and illustrate the doubly robust approach through simulated case studies. By simplifying the methodology and offering practical coding examples, we intend to make doubly robust learning accessible to researchers and practitioners in data science and statistics.