Causal Inference with Machine Learning

Session Submission Type: Workshop

Abstract

Causal Inference with Machine Learning

A critical limitation of conventional methods is that they typically operate under assumptions of linear effects and correct model specification (e.g., all non-linearities and n-way interactions are known and included). Recent research in machine learning has relaxed such limitations by developing data adaptive methods that empirically construct these relationships. For these reasons, machine learning methods hold significant potential to relax key assumptions when estimating causal effects. In this workshop, we introduce the targeted learning framework. The targeted learning framework integrates machine learning methods and the potential outcomes framework to develop causal inferences. The result is a highly flexible and data adaptive doubly robust estimator. Examples on education in R will be used to introduce the framework and provide practical applications.

Policy Area

• Workshops

Moderator

• Ben Kelcey, University of Cincinnati

Organizer

• Ben Kelcey, University of Cincinnati

Speakers

• Ben Kelcey, University of Cincinnati

• Amota Ataneka, University of Cincinnati