基于R-Vine Copula模型的情景模拟研究
SCENARIO SIMULATION OF R-VINE COPULA MODEL
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.
Scenario simulation is a common method to determine the importance of a variable in a multi variable system and accurately modeling the dependency between variables is crucial to scenario simulation. In this paper, the scenerio simulation method based on more flexible R-vine copula is proposed, from which one can simulate the responses of other individuals and the whole system to an individual when it is in the upper and lower tail extreme situations. Through simulation studies and analysis of a sample system containing five German company dataset, the systemically important enterprise in different market quotation are successfully found out. The empirical results are consistent with the reality, which proves that the method proposed in this paper is reliable and can be widely used for identification of systemically important institutions and macro prudential supervision of risk contagion in the field of financial management.
/
〈 | 〉 |