B1737
Title: A double machine learning approach to combining experimental and observational data
Authors: Alexander Volfovsky - Duke University (United States) [presenting]
Abstract: Experimental and observational studies often lack validity due to untestable assumptions. A double machine learning approach is proposed to combine experimental and observational studies, allowing practitioners to test for assumption violations and estimate treatment effects consistently. The framework tests for violations of external validity and ignorability under milder assumptions. When only one assumption is violated, semiparametrically efficient treatment effect estimators are provided. However, the no-free-lunch theorem highlights the necessity of accurately identifying the violated assumption for consistent treatment effect estimation. The applicability of the approach is demonstrated in three real-world case studies, highlighting its relevance for practical settings.
