主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2024, Vol. 32 ›› Issue (12): 173-182.doi: 10.16381/j.cnki.issn1003-207x.2022.1546

• • 上一篇    下一篇

基于多层网络结构的行业间风险联动机制研究

沈虹1(), 张晨曜1,2, 刘晓星2   

  1. 1.扬州大学商学院,江苏 扬州 225127
    2.东南大学经济管理学院,江苏 南京 211189
  • 收稿日期:2022-07-15 修回日期:2022-12-29 出版日期:2024-12-25 发布日期:2025-01-02
  • 通讯作者: 沈虹 E-mail:shenhong@yzu.edu.cn
  • 基金资助:
    国家自然科学基金项目(72173018);江苏省自然科学基金项目(BK20170515)

Analysis of Risk Spillover Characteristics and Mechanism among Industries: Evidence from Multilayer Network

Hong Shen1(), Chenyao Zhang1,2, Xiaoxing Liu2   

  1. 1.School of Business,Yangzhou University,Yangzhou 225127,China
    2.School of Economics and Management,Southeast University,Nanjing 211189,China
  • Received:2022-07-15 Revised:2022-12-29 Online:2024-12-25 Published:2025-01-02
  • Contact: Hong Shen E-mail:shenhong@yzu.edu.cn

摘要:

本文通过构建DCC-t-Copula-CoVaR模型和多层网络结构,有效测度了国内各行业间的风险联动效应以及风险传播路径。同时,结合网络结构特征,使用混频回归法对行业间的风险传染因素展开深入研究。研究发现:从风险传导方面,行业间风险呈现出多属性、多类型的联动关系,多层网络理论能够较好地刻画行业间的复杂关联性;单个行业风险难以造成连锁反应,但行业间风险共振会导致重大系统性风险;重大风险事件冲击下,行业风险网络多中心发展,传染路径多元化演变。影响因素方面,重要行业在风险传染中起关键作用,类似“系统重要性角色”,且重要行业的紧邻行业同样值得关注;网络中关联度高、接近中心的行业,既容易受到风险冲击,又具有较强的风险分散能力,体现出“稳定又脆弱”的特征。在此基础上,本文结合当前经济稳中求进的政策背景,提出防范化解行业间重大系统性风险的对策建议,以期为增强宏观经济韧性,实现经济高质量发展提供理论依据。

关键词: 系统性风险, 风险溢出, 多层网络, 混频回归

Abstract:

The report of the 20th National Congress of the Communist Party of China pointed out that “we should strengthen and improve modern financial supervision, strengthen the financial stability guarantee system, and bring all kinds of financial activities under supervision according to law”. In the time dimension, the accelerator effect between different industries forms a risk feedback mechanism; in the spatial dimension, different industries have amplified the systematic risk through the formation of complex financial associations through the role of the market. As a result, studying the risk contagion mechanism among industries is of great significance for preventing and resolving major systemic risks and maintaining financial stability.The CSI All Share Index is taken as object, DCC-t-Copula-CoVaR model is used to measure systematic risk spillovers and absorption levels as the explained variables, a multilayer network is constructed, key structural indicators are taken as the explanatory variables, and MIDAS method is used to study the risk linkage mechanism.Consequently, suggestions are put forward to prevent systemic risks among industries:The regulatory level should consider the time-varying characteristics of the risk network and describe the dynamic evolution mechanism of systematic risk. A macro prudential risk monitoring mechanism of “too relevant to fail” based on the perspective of multi-layer networks should also be designed. To prevent and defuse systemic risks, the Party's leadership over financial should be strengthened, pushing our economy towards high-quality development.A reference model and basis for establishing a cross industry risk linkage monitoring system is provided. Through effective monitoring of inter industry risk linkage, the ability to withstand extreme risk shocks is improved, so as to increase the resilience of the macro-economy, make progress while maintaining stability, and achieve high-quality development of the industry.It is found that from the perspective of risk transmission, multilayer network theory can better describe the complex correlation among industries. It is difficult for a single industry to cause a chain reaction,but risk resonance among industries will lead to major systemic risks. Industry risk network polycentric development and diversification of contagion paths evolve under the impact of major risk events. From the aspect of influencing factors: The key role of important industries in risk contagion is similar to the "systemically important role",and the adjacent industries of important industries also deserve attention. The industries which have high relevance and close to the center in the network reflect the characteristics of both stable and fragile.

Key words: systemic risk, risk spillover, multilayer networks, MIDAS

中图分类号: