引入稳定分布对沪深两市指数数据进行检验,结果表明两指数具有“尖峰厚尾”的分形特征,在此基础上建立了DaR类风险测度,实证研究表明两指数跌幅时间序列存在协同跌幅共线性效应.其次,给出了蒙特卡洛稳定分布和正态分布模拟下的两类风险测度估计值,建立了离差率模型,结果表明稳定分布下的风险度量适合投资者进行风险管理.最后,研究了不同跟踪时间窗口下的风险测度指标MDD.投资者和风险管理人员不仅要关注VaR类风险,更要警惕DaR类风险指标.
The paper uses the Alpha-stable distribution to examine the data of Shanghai composite index and Shenzhen component index. The results show that both of them have the fractal characteristic of "leptokurtic and heavy tails". We establish DaR-type risk measures. The empirical study demonstrates that series between the two indexes have collinearity to some extent in terms of drawdown. Furthermore, using the Alpha-stable parameters of the two indexes, we give the VaR-type and DaR-type risk measures estimation in the Monte Carlo Alpha-stable and normal simulations, and we construct the model of bias. Finally, this paper emphasizes on MDD with different tracking time. As investors and risk managers, we should focus on not only VaR-type risk measures, but also DaR-type risk indexes.