Estimating Treatment Effect Heterogeneity with Varying Coefficients and Bayesian Tree Ensembles

• In Event: Observational Causal Inference

Brief Overview

We introduce a new framework for estimating heterogeneous treatment effects in difference-in-difference designs by using Bayesian Additive Regression Trees. We apply the method to understand how increasing workloads affect the judicial system.

Authors

• Saloni Bhogale, University of Wisconsin–Madison

• Sameer Deshpande, University of Wisconsin–Madison