Detecting Treatment Interference under the K-Nearest-Neighbors Interference Model

TL;DR

This paper introduces a K-nearest neighbors interference model for treatment effects, providing a new approach to select focal units that enhances the power of detecting interference compared to existing methods.

Contribution

The paper proposes a novel K-nearest neighbors interference model and offers a method for selecting focal units that improves the detection of treatment interference.

Findings

Focal unit selection under the model increases test power.

Simulation results show improved detection of interference.

Existing methods are less effective than the proposed approach.

Abstract

We propose a model of treatment interference where the response of a unit depends only on its treatment status and the statuses of units within its K-neighborhood. Current methods for detecting interference include carefully designed randomized experiments and conditional randomization tests on a set of focal units. We give guidance on how to choose focal units under this model of interference. We then conduct a simulation study to evaluate the efficacy of existing methods for detecting network interference. We show that this choice of focal units leads to powerful tests of treatment interference which outperform current experimental methods.