Population Interference in Panel Experiments

62 Pages Posted: 11 Mar 2021

See all articles by Kevin Wu Han

Kevin Wu Han

Stanford University, Department of Statistics

Iavor Bojinov

Harvard University - Technology & Operations Management Unit; Harvard University, Department of Statistics, Students

Guillaume Basse

Stanford University - Department of Statistics

Date Written: March 1, 2021

Abstract

The phenomenon of population interference, where a treatment assigned to one experimental unit affects another experimental unit's outcome, has received consider- able attention in standard randomized experiments. The complications produced by population interference in this setting are now readily recognized, and partial remedies are well known. Much less understood is the impact of population interference in panel experiments where treatment is sequentially randomized in the population, and the outcomes are observed at each time step. This paper proposes a general frame- work for studying population interference in panel experiments and presents new finite population estimation and inference results. Our findings suggest that, under mild assumptions, the addition of a temporal dimension to an experiment alleviates some of the challenges of population interference for certain estimands. In contrast, we show that the presence of carryover effects | that is, when past treatments may affect future outcomes | exacerbates the problem. Revisiting the special case of standard experiments with population interference, we prove a central limit theorem under weaker conditions than previous results in the literature and highlight the trade-off between flexibility in the design and the interference structure.

Keywords: Finite Population, Potential Outcomes, Dynamic Causal Effects

Suggested Citation

Wu Han, Kevin and Bojinov, Iavor and Basse, Guillaume, Population Interference in Panel Experiments (March 1, 2021). Harvard Business School Technology & Operations Mgt. Unit Working Paper No. 21-100, Available at SSRN: https://ssrn.com/abstract=3802304 or http://dx.doi.org/10.2139/ssrn.3802304

Kevin Wu Han

Stanford University, Department of Statistics

Stanford, CA 94305
United States

Iavor Bojinov (Contact Author)

Harvard University - Technology & Operations Management Unit ( email )

Boston, MA 02163
United States

Harvard University, Department of Statistics, Students ( email )

Cambridge, MA
United States

Guillaume Basse

Stanford University - Department of Statistics ( email )

Stanford, CA 94305
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
63
Abstract Views
212
rank
412,644
PlumX Metrics