Saved in:
Bibliographic Details
Main Author: Crippa, Federico
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.11767
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915228437970944
author Crippa, Federico
author_facet Crippa, Federico
contents Threshold policies are decision rules that assign treatments based on whether an observable characteristic exceeds a certain threshold. They are widespread across multiple domains, including welfare programs, taxation, and clinical medicine. This paper examines the problem of designing threshold policies using experimental data, when the goal is to maximize the population welfare. First, I characterize the regret - a measure of policy optimality - of the Empirical Welfare Maximizer (EWM) policy, popular in the literature. Next, I introduce the Smoothed Welfare Maximizer (SWM) policy, which improves the EWM's regret convergence rate under an additional smoothness condition. The two policies are compared by studying how differently their regrets depend on the population distribution, and investigating their finite sample performances through Monte Carlo simulations. In many contexts, the SWM policy guarantees larger welfare than the EWM. An empirical illustration demonstrates how the treatment recommendations of the two policies may differ in practice.
format Preprint
id arxiv_https___arxiv_org_abs_2404_11767
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Regret Analysis in Threshold Policy Design
Crippa, Federico
Econometrics
Threshold policies are decision rules that assign treatments based on whether an observable characteristic exceeds a certain threshold. They are widespread across multiple domains, including welfare programs, taxation, and clinical medicine. This paper examines the problem of designing threshold policies using experimental data, when the goal is to maximize the population welfare. First, I characterize the regret - a measure of policy optimality - of the Empirical Welfare Maximizer (EWM) policy, popular in the literature. Next, I introduce the Smoothed Welfare Maximizer (SWM) policy, which improves the EWM's regret convergence rate under an additional smoothness condition. The two policies are compared by studying how differently their regrets depend on the population distribution, and investigating their finite sample performances through Monte Carlo simulations. In many contexts, the SWM policy guarantees larger welfare than the EWM. An empirical illustration demonstrates how the treatment recommendations of the two policies may differ in practice.
title Regret Analysis in Threshold Policy Design
topic Econometrics
url https://arxiv.org/abs/2404.11767