论文部分内容阅读
相关性分析是多变量分析中的一个中心问题,而压力情景模拟则是确定某变量在多变量系统中重要性的常用方法.文章针对灵活刻画多变量相依结构的R-vine copula模型,提出了R-vine结构下的情景模拟算法,并以德国五公司收益率序列为样本系统,发现了系统内个体相依结构,模拟了个体在上、下尾极值情景下,其他个体及整个系统的响应情况,发现了不同行情下的系统重要性企业,实证结果与现实基本一致.所提出的算法可用于金融管理领域的风险传染、系统重要性机构的识别以及宏观审慎监管等方面的研究.
Correlation analysis is a central issue in multivariate analysis, and stress scenario simulation is a common method to determine the importance of a variable in a multivariable system.For the R-vine copula model that describes multivariable dependent structures flexibly, R-vine structure of the scene simulation algorithm, and the five German company yield sequence as a sample system, found that the individual dependencies in the system structure, the simulation of individuals in the upper and lower tail scenarios, other individuals and the entire system response We found the systemically important enterprises under different market conditions, and the empirical results are basically the same as the reality.The proposed algorithm can be used for the research of risk contagion in the financial management field, the identification of the system importance institution and macroprudential supervision.