Rate-optimal cluster-randomized designs for spatial interference

We consider a potential outcomes model in which interference may be present between any two units but the extent of interference diminishes with spatial distance. The causal estimand is the global average treatment effect, which compares counterfactual outcomes when all units are treated to those when none are. We study a class of designs in which space is partitioned into clusters that are randomized into treatment and control. For each design, we estimate the treatment effect using a Horvitz-Thompson estimator that compares the average outcomes of units with all neighbors treated to units with no treated neighbors, where the neighborhood radius is of the same order as the cluster size dictated by the design. We derive the estimator’s rate of convergence as a function of the design and degree of interference and use this to obtain estimator-design pairs that achieve near-optimal rates of convergence under relatively minimal assumptions on interference. We prove that the estimators are asymptotically normal and provide a variance estimator. For practical implementation of the designs, we suggest partitioning space using clustering algorithms. only be directly observed in the data under an extreme design that assigns all units to the same treatment arm, which would necessarily preclude observation of the other counterfactual. Common designs used in the literature, including those studied here, assign different units to different treatment arms, so neither average is directly observed in the data. Nonetheless, we show that asymptotic inference on θ n is possible for a class of cluster-randomized designs under spatial interference where the degree of interference diminishes with distance.