Causal Inference on Networks under Misspecified Exposure Mappings: A Partial Identification Framework

2026-02-03

AI Summary

This study addresses the pervasive issue of exposure mapping misspecification in network causal inference, which often leads to severe bias in estimating direct and spillover effects. For the first time, it introduces causal sensitivity analysis into this setting by proposing a partial identification framework that constructs sharp bounds on causal effects under misspecification, thereby enabling robustness assessment of network intervention effects. Focusing on three canonical exposure specifications—weighted averages, threshold mappings, and higher-order truncated interference—the work develops orthogonal estimators that ensure the resulting bounds are not only informative but also statistically efficient and valid. This approach provides a verifiable, robust, and computationally tractable toolkit for causal inference under network interference.

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
• Social Networks and Social Media: Social media analysis through the lenses of networks
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Estimating treatment effects in networks is challenging, as each potential outcome depends on the treatments of all other nodes in the network. To overcome this difficulty, existing methods typically impose an exposure mapping that compresses the treatment assignments in the network into a low-dimensional summary. However, if this mapping is misspecified, standard estimators for direct and spillover effects can be severely biased. We propose a novel partial identification framework for causal inference on networks to assess the robustness of treatment effects under misspecifications of the exposure mapping. Specifically, we derive sharp upper and lower bounds on direct and spillover effects under such misspecifications. As such, our framework presents a novel application of causal sensitivity analysis to exposure mappings. We instantiate our framework for three canonical exposure settings widely used in practice: (i) weighted means of the neighborhood treatments, (ii) threshold-based exposure mappings, and (iii) truncated neighborhood interference in the presence of higher-order spillovers. Furthermore, we develop orthogonal estimators for these bounds and prove that the resulting bound estimates are valid, sharp, and efficient. Our experiments show the bounds remain informative and provide reliable conclusions under misspecification of exposure mappings.

Problem

• causal inference
• networks
• exposure mapping
• misspecification
• spillover effects

Innovation

• partial identification
• exposure mapping misspecification
• network causal inference
• spillover effects
• orthogonal estimation