This paper integrates the Elastic Net-TVP-VAR-BK framework and constructs a high-dimensional multilayer frequency-domain network, including short-, medium-, and long-term layers, to investigate extreme risk spillovers among 56 industries in the Chinese stock market. We examine the topology of the multilayer network at the system, cross-sector, and industry levels, as well as from both static and dynamic perspectives. Using daily data on 56 industry indices from 1 March 2007 to 30 September 2024, our empirical results show that: (1) All multilayer network topologies, including edge structures, node characteristics, and spillover strengths, exhibit significant frequency heterogeneity, and the dynamic topology of the three-layer network shows fluctuations and directional differences during critical periods. (2) In most periods, the short-term layer exhibits stronger average spillover intensity and denser inter-industry linkages, suggesting that short-horizon risk transmission plays a more prominent role in rapid contagion. However, the medium- and long-term layers remain important for identifying persistent and structural risk transmission. (3) At the industry level, capital markets and textiles, apparel, and luxury goods within the short-term layer, food products, household products, and road and rail in the medium-term layer as well as construction and engineering, industrial conglomerates, trading companies and distributors, metals and mining, and distributors in the long-term layer, all demonstrate high cross-industry systemic importance and total systemic importance, thereby establishing themselves as key nodes within their respective frequency domains. The findings provide theoretical support for policymakers in formulating strategies to address market risks and offer important references for investors in asset allocation and risk management decisions.
Yi et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: