Title: An Improved Double Robust Approach for Precision Medicine
Abstract: Estimating the Average Treatment Effect is a key problem in both causal inference and precision medicine. The double robust method has been widely used in this setting. In this talk, an energy-distance-based approach is proposed in this paper to estimate the treatment effect. The new approach avoids the estimation of the typical propensity score, but can also achieve double robustness. Theoretical justification and examples will be used to show the usefulness of this new approach.