我国金融体系跨部门系统性风险特点--基于广义方差分解的双向溢出网络  

Measuring Systemic Risk Spillover Network in China’s Financial Market Based on Generalized FEVD Model

作　　者：吴静 WU Jing(Antai College of Economics&Management,Shanghai Jiao Tong University,Shanghai 200030,China)

机构地区：[1]上海交通大学安泰经济与管理学院,上海200030

出　　处：《上海管理科学》2022年第5期62-66,共5页Shanghai Management Science

摘　　要：为了研究我国金融部门间系统性风险溢出的整体分布和动态变化,文章借助连通度指标来衡量系统性风险溢出,这一指标更加关注整体性,符合系统性风险全面传染的特点。文章用2007-2019年中国金融部门内共28家上市公司的股价日内波动率,依据Diebold和Yilmaz(2014)提出的广义方差法,建立了金融市场间的双向波动溢出网络。首先,静态结果表明样本公司在部门内部和部门之间的联系都比较紧密,主营业务范围对跨部门联系起到了决定性的作用。其次,银行是市场的核心风险输出者(C_(insurance←bank)=182.3,C_(security←bank)=79,C_(trust&others←bank)=261.8),保险公司具有重要的影响力。最后,动态总连通度指数为63.11~95.21,总连通度可以作为市场的指标,指示具有不对称性。To measure overall distribution and dynamic change of systemic risk spillovers among financial sectors in China,this paper establishes a connectedness network.We establish the volatility spillover network created by Diebold and Yilmaz(2014),including bankers,insurers,securities,trust&others,based on estimated daily volatility of 28 companies from different department of Chinese financial market from 2007-2019.First,results show companies are connected within and cross the department,and the main business is a decisive factor.Second,banks are the core candidates in the market(C_(insurance←bank)=182.3,C_(security←bank)=79,C_(trust&others←bank)=261.8),insurers have importance influence as well,they form the two main circles of the system,the rest companies are pulled between the two circles.Third,the dynamic connectedness can function as an indicator of the market,and the indication effect is asymmetric.

关 键 词：系统性风险 广义方差分解 连通矩阵 有向网络 

分 类 号：F832[经济管理—金融学]