Optimizing Cluster-based Randomized Experiments under Monotonicity

Citations Over TimeTop 10% of 2018 papers

Abstract

Cluster-based randomized experiments are popular designs for mitigating the bias of standard estimators when interference is present and classical causal inference and experimental design assumptions (such as SUTVA or ITR) do not hold. Without an exact knowledge of the interference structure, it can be challenging to understand which partitioning of the experimental units is optimal to minimize the estimation bias. In the paper, we introduce a monotonicity condition under which a novel two-stage experimental design allows us to determine which of two cluster-based designs yields the least biased estimator. We then consider the setting of online advertising auctions and show that reserve price experiments satisfy the monotonicity condition and the proposed framework and methodology apply. We validate our findings on an advertising auction dataset.

Related Papers

• → Designing and Analyzing Randomized Experiments: Application to a Japanese Election Survey Experiment(2007)103 cited
• → Natural Experiments: Missed Opportunities for Causal Inference in Psychology(2023)12 cited
• → Causal Inference Preliminary(2023)1 cited
• → A Survey on Causal Inference(2020)24 cited
• → Design-based inference for generalized network experiments with stochastic interventions(2023)