Assessing systemic importance using multilayer dynamic networks: Evidence from China's stock market

Abstract

This study develops a multilayer dynamic network framework to evaluate the systemic importance of 348 firms listed in China's A-share market over the period 2010–2021. By employing the maximum mutual information coefficient (MIC), the model captures both linear and nonlinear interdependencies, integrating firm-specific tail risk indicators and trading-based metrics. Topological analysis of the network, including connectivity, clustering, and centrality measures, reveals structural drivers of systemic risk propagation. The results show that firms with high centrality and interconnectedness disproportionately amplify systemic vulnerabilities, underscoring their critical roles in financial stability. The multilayer dynamic framework significantly enhances the precision of systemic risk assessment compared to traditional single-layer models. This study contributes to systemic risk literature by extending advanced network methodologies to emerging markets and offers actionable insights for policymakers and regulators to design effective risk mitigation strategies.