A Nonparametric Test for Cross-Unit Spillovers

2026-07-31

AI Summary

This study addresses the challenge of effectively testing spillover effects under nonlinear dependence structures across units. It proposes a nonparametric test that requires estimation only under the null hypothesis of no spillovers, enabling detection of spillovers induced by peers’ attributes, outcomes, or both. The method establishes, for the first time, a unified testing framework applicable to diverse interaction structures, without requiring explicit modeling of the spillover mechanism and featuring a standard normal asymptotic distribution. By integrating nonparametric inference with a flexible data-generating process, the approach enhances robustness and applicability. Empirical applications across four settings yield spillover conclusions markedly different from those of existing methods, underscoring its effectiveness and practical utility.

Abstract

Cross-unit dependence is pervasive in empirical applications and complicates econometric inference, especially when spillovers operate in nonlinear ways. We propose a novel nonparametric test for cross-unit spillovers that may operate through peers' attributes, peers' outcomes, or both. The test is straightforward to implement, as it requires only estimation under the null hypothesis of no cross-unit spillovers, and is shown to have a convenient asymptotic standard normal distribution. It is also versatile, accommodating data generated by a wide range of interaction structures. We present four empirical illustrations showing that the proposed test can yield substantively different conclusions about the presence of cross-unit spillovers than existing approaches.

Problem

• cross-unit spillovers
• nonparametric test
• peer effects
• econometric inference
• nonlinear dependence

Innovation

• nonparametric test
• cross-unit spillovers
• peer effects
• asymptotic normality
• interaction structures