Design-based causal inference in bipartite experiments

2025-01-16

AI Summary

This paper addresses causal inference under interference in bipartite experiments—where treatment and outcome units are distinct—under weak modeling assumptions. We develop a design-based causal framework grounded in experimental design principles. First, we introduce a foundational causal model on bipartite graphs, leveraging graph sparsity to characterize interference patterns. Second, we propose an unbiased and consistent estimator for the average treatment effect (ATE). Third, we establish a central limit theorem and derive a conservative variance estimator. Finally, we incorporate covariate adjustment to improve estimation efficiency. Unlike conventional approaches, our method avoids strong parametric assumptions about the data-generating process, enabling robust statistical inference. It significantly enhances estimation accuracy and broad applicability, providing an interpretable, reproducible causal analysis tool for canonical bipartite settings such as platform economies and recommender systems.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Bipartite experiments arise in various fields, in which the treatments are randomized over one set of units, while the outcomes are measured over another separate set of units. However, existing methods often rely on strong model assumptions about the data-generating process. Under the potential outcomes formulation, we explore design-based causal inference in bipartite experiments under weak assumptions by leveraging the sparsity structure of the bipartite graph that connects the treatment units and outcome units. We make several contributions. First, we formulate the causal inference problem under the design-based framework that can account for the bipartite interference. Second, we propose a consistent point estimator for the total treatment effect, a policy-relevant parameter that measures the difference in the outcome means if all treatment units receive the treatment or control. Third, we establish a central limit theorem for the estimator and propose a conservative variance estimator for statistical inference. Fourth, we discuss a covariate adjustment strategy to enhance estimation efficiency.

Problem

• Causal inference in bipartite experiments with weak assumptions
• Estimating total treatment effect under bipartite interference
• Developing design-based framework for sparse bipartite graphs

Innovation

• Design-based causal inference under weak assumptions
• Consistent estimator for total treatment effect
• Covariate adjustment for enhanced efficiency