Primary Submission Category: Heterogeneous Treatment Effects
Optimal decision rules based on algorithms and human intuition
Authors: Mats Stensrud, Aaron Sarvet,
Presenting Author: Mats Stensrud*
Health care providers desire to implement decision rules that, when applied to individuals in the population of interest, yield the best possible outcomes. For example, the current focus on precision medicine reflects the search for individualized treatment decisions, adapted to a patient’s characteristics. In this presentation, I will introduce superoptimal regimes, which are guaranteed to outperform conventional optimal regimes. Importantly, identification of superoptimal regimes and their values require exactly the same assumptions as identification of conventional optimal regimes in several common settings. To illustrate the utility of superoptimal regimes, we derive new identification and estimation results, including a semi-parametric efficient estimator, in a common instrumental variable setting. These results are used to study two data examples that have appeared in the optimal regimes literature, illustrating that the superoptimal regimes perform better than conventional optimal regimes.