Behavioral, Econometrics and Theory Seminar Series Presents: Kevin Chen
October 23 @ 1:40 pm
Event Series
(See All)
Economics Behavioral, Econometrics, & Theory Seminar
Date: Thursday, October 23, 2025
Time: 1:40-3:00 p.m.
Location: Engineering 2, Rm 499
Speaker: Kevin Chen
Title: Assistant Professor of Economics
Affiliation: Stanford University
Host: Michael Leung
Seminar title: Compound Selection Decisions: An Almost SURE Approach
ABSTRACT: This paper proposes methods for producing compound selection decisions in a Gaussian sequence model. Given unknown, fixed parameters µ_{1:n} and known σ_{1:n} with observations Yᵢ ∼ 𝒩(μᵢ, σᵢ²), the aim is to select a subset of units S to maximize utility Σ_{i∈S}(μᵢ − Kᵢ) for known costs Kᵢ. Inspired by Stein’s unbiased risk estimate (SURE), we introduce an almost unbiased estimator, ASSURE, for the expected utility of a proposed decision rule. ASSURE allows a user to choose a welfare-maximizing rule from a pre-specified class by optimizing the estimated welfare, thereby producing selection decisions that borrow strength across noisy estimates. We show that ASSURE yields decision rules that are asymptotically no worse than the optimal but infeasible rule in the pre-specified class. We apply ASSURE to p-value decision procedures in A/B testing, selecting Census tracts for economic opportunity, and identifying discriminating firms.
