论文部分内容阅读
股市诸多行业风险之间存在着波动相依性,集成计量多维风险对投资决策意义重大。藤Copula是Copula函数高维化拓展的一个方向,其动态化是新的研究前沿。将极值理论的GPD模型和高维动态C藤Copula方法结合起来研究沪深300指数中地产、基建、银行和运输四个行业风险,能够有效描述尾部极值形态,突出关键变量的作用。再运用动态Pair-Copula分解,刻画高维行业风险变量间的动态关系,以仿真出动态集成风险变量VaR序列。VaR计算结果通过了回溯检验和稳定性测试,表明高维动态C藤Copula模型可以作为风险集成计量的一种新的有效方法。
There are volatility dependencies among many industry risks in the stock market. The integrated measurement of multi-dimensional risks is of great significance to investment decisions. Copula Copula is a direction of high-dimensional expansion of Copula function, its dynamic is a new research frontier. Combining the extremum theory GPD model with the high-dimensional dynamic C Rat Copula method to study the risk of the four industries in real estate, infrastructure, banking and transportation in the CSI 300 Index can effectively describe the tail extremum morphology and highlight the role of key variables. Dynamic Pair-Copula decomposition is also used to characterize the dynamic relationships between high-dimensional industry risk variables to simulate VaR sequences of dynamic integrated risk variables. The results of VaR pass the backtesting and stability test, which shows that the high dimensional dynamic C Rattle Copula model can be used as a new effective method for risk integration measurement.