Estimating Peer Direct and Indirect Effects in Observational Network Data

TL;DR

This paper introduces a comprehensive method for estimating both direct and indirect peer effects in observational network data using graph neural networks enhanced with HSIC, improving causal inference accuracy.

Contribution

It proposes a novel framework that captures multiple peer effects and employs attention and HSIC within GNNs to improve robustness and interpretability.

Findings

Effective estimation of peer effects demonstrated on semi-synthetic datasets.

The approach outperforms existing methods in accuracy and robustness.

Theoretical insights support broader applications in social networks and epidemiology.

Abstract

Estimating causal effects is crucial for decision-makers in many applications, but it is particularly challenging with observational network data due to peer interactions. Many algorithms have been proposed to estimate causal effects involving network data, particularly peer effects, but they often overlook the variety of peer effects. To address this issue, we propose a general setting which considers both peer direct effects and peer indirect effects, and the effect of an individual's own treatment, and provide identification conditions of these causal effects and proofs. To estimate these causal effects, we utilize attention mechanisms to distinguish the influences of different neighbors and explore high-order neighbor effects through multi-layer graph neural networks (GNNs). Additionally, to control the dependency between node features and representations, we incorporate the Hilbert-Schmidt Independence Criterion (HSIC) into the GNN, fully utilizing the structural information of the graph, to enhance the robustness and accuracy of the model. Extensive experiments on two semi-synthetic datasets confirm the effectiveness of our approach. Our theoretical findings have the potential to improve intervention strategies in networked systems, with applications in areas such as social networks and epidemiology.