Centrality-Based Heuristics in Risk Management: A Network-agent Dynamics for Analyzing Systemic Failure

Abstract

This study scrutinized the mechanisms for mitigating systemic risk in scale-free networks, modeling on the functions of memorized capital, social learning, and centrality-based heuristics. We deployed a network-agent dynamic to examine the distribution of vulnerabilities and explored how node centrality impacts decisions related to protective investment. Our mechanical approach encompassed advanced computational techniques and interactive simulations, enabling us to monitor the relevant state variables effectively. By focusing on the dynamics of individual protective investment, we observed that nodes with higher centrality are inclined to invest more significantly in protection. These outcomes provide crucial understandings of the risk propagation and the tactics to mitigate such risks, which underscores the significance of strategic cooperation and effective regulation in bolstering network resilience and stability.