Abstract:Variance reduction for causal inference in the presence of network interference is often achieved through either outcome modeling, which is typically analyzed under unit-randomized Bernoulli designs, or clustered experimental designs, which are typically analyzed without strong parametric assumptions. In this work, we study the intersection of these two approaches and consider the problem of estimation in low-order outcome models using data from a general experimental design. Our contributions are threefold. First, we present an estimator of the total treatment effect (also called the global average treatment effect) in a low-degree outcome model when the data are collected under general experimental designs, generalizing previous results for Bernoulli designs. We refer to this estimator as the pseudoinverse estimator and give bounds on its bias and variance in terms of properties of the experimental design. Second, we evaluate these bounds for the case of cluster randomized designs with both Bernoulli and complete randomization. For clustered Bernoulli randomization, we find that our estimator is always unbiased and that its variance scales like the smaller of the variance obtained from a low-order assumption and the variance obtained from cluster randomization, showing that combining these variance reduction strategies is preferable to using either individually. For clustered complete randomization, we find a notable bias-variance trade-off mediated by specific features of the clustering. Third, when choosing a clustered experimental design, our bounds can be used to select a clustering from a set of candidate clusterings. Across a range of graphs and clustering algorithms, we show that our method consistently selects clusterings that perform well on a range of response models, suggesting that our bounds are useful to practitioners.

Optimality of Circular Neighbor-Balanced Designs for Total Effects with Autoregressive Correlated Observations

Identification Of Universally Optimal Circular Designs For The Interference Model

Circular Neighbor-Balanced Designs Universally Optimal for Total Effects

Optimal designs for crossover model with partial interactions

Optimal designs for the proportional interference model

Optimized Covariance Design for AB Test on Social Network under Interference

The Optimality of Blocking Designs in Equally and Unequally Allocated Randomized Experiments with General Response

Optimality of Pairwise Blocked Definitive Screening Designs

Low-order outcomes and clustered designs: combining design and analysis for causal inference under network interference

Exploiting Neighborhood Interference with Low Order Interactions under Unit Randomized Design

Copula-based robust optimal block designs

Experimental designs for controlling the correlation of estimators in two parameter models

On optimality of designs with three distinct eigenvalues

Optimal Bayesian stepped-wedge cluster randomised trial designs for binary outcome data

Optimal BIBD-extended designs

Contrasting Multiple Social Network Autocorrelations for Binary Outcomes, With Applications To Technology Adoption

Some Notes on Constructions of Binary Sequences with Optimal Autocorrelation

Common Correlated Effects Estimation of Nonlinear Panel Data Models

Robustness of response-adaptive randomization

On the Total Least-Squares Estimation for Autoregressive Model

E- and R-optimality of block designs for treatment-control comparisons