Saved in:
Bibliographic Details
Main Authors: Jia, Su, Kallus, Nathan, Yu, Christina Lee
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2312.15574
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916663282106368
author Jia, Su
Kallus, Nathan
Yu, Christina Lee
author_facet Jia, Su
Kallus, Nathan
Yu, Christina Lee
contents We consider experimentation in the presence of non-stationarity, inter-unit (spatial) interference, and carry-over effects (temporal interference), where we wish to estimate the global average treatment effect (GATE), the difference between average outcomes having exposed all units at all times to treatment or to control. We suppose spatial interference is described by a graph, where a unit's outcome depends on its neighborhood's treatments, and that temporal interference is described by an MDP, where the transition kernel under either treatment (action) satisfies a rapid mixing condition. We propose a clustered switchback design, where units are grouped into clusters and time steps are grouped into blocks, and each whole cluster-block combination is assigned a single random treatment. Under this design, we show that for graphs that admit good clustering, a truncated Horvitz-Thompson estimator achieves a $\tilde O(1/NT)$ mean squared error (MSE), matching the lower bound up to logarithmic terms for sparse graphs. Our results simultaneously generalize the results from \citet{hu2022switchback,ugander2013graph} and \citet{leung2022rate}. Simulation studies validate the favorable performance of our approach.
format Preprint
id arxiv_https___arxiv_org_abs_2312_15574
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Clustered Switchback Designs for Experimentation Under Spatio-temporal Interference
Jia, Su
Kallus, Nathan
Yu, Christina Lee
Statistics Theory
Machine Learning
We consider experimentation in the presence of non-stationarity, inter-unit (spatial) interference, and carry-over effects (temporal interference), where we wish to estimate the global average treatment effect (GATE), the difference between average outcomes having exposed all units at all times to treatment or to control. We suppose spatial interference is described by a graph, where a unit's outcome depends on its neighborhood's treatments, and that temporal interference is described by an MDP, where the transition kernel under either treatment (action) satisfies a rapid mixing condition. We propose a clustered switchback design, where units are grouped into clusters and time steps are grouped into blocks, and each whole cluster-block combination is assigned a single random treatment. Under this design, we show that for graphs that admit good clustering, a truncated Horvitz-Thompson estimator achieves a $\tilde O(1/NT)$ mean squared error (MSE), matching the lower bound up to logarithmic terms for sparse graphs. Our results simultaneously generalize the results from \citet{hu2022switchback,ugander2013graph} and \citet{leung2022rate}. Simulation studies validate the favorable performance of our approach.
title Clustered Switchback Designs for Experimentation Under Spatio-temporal Interference
topic Statistics Theory
Machine Learning
url https://arxiv.org/abs/2312.15574