Anja Sautmann, J-PAL

Sala 209

“Adaptive Treatment Assignment in Experiments for Policy Choice”

Coautoreado con Maximilian Kasy

Abstract: The goal of many experiments is to inform the choice between different policies. However, standard experimental designs are geared toward point estimation and hypothesis testing. We consider the problem of treatment assignment in an experiment with several cross-sectional waves where the goal is to choose among a set of possible policies (treatments) for large-scale implementation. We show that optimal experimental designs learn from earlier waves by assigning more experimental units to the better-performing treatments in later waves. We discuss a computationally tractable approximation of the optimal design, based on a modification of Thompson sampling. Calibrated simulations and theoretical results demonstrate improvements in welfare, relative to conventional designs as well as standard Thompson sampling. Our setting is related to but different from multi- armed bandit settings. The focus on the highest-performing policies is not driven by an “exploitation” motive, but by optimal learning about the best policy choice.