论文部分内容阅读
考虑中国股市指数收益率分布和波动的非对称性结构,采用偏t分布拟合收益率的有偏分布形态,利用RS-捕捉波动率的杠杆效应,并构建ARFIMA-GARCH和SKST-RS-模型分别预测RS-和刻画收益率波动的动态结构,进而改进VaR和ES并测度卖空限制市场的下侧风险。通过Kupiec LR和动态分位数检验,实证分析了ES和VaR的风险管理效果。结果表明:基于日内高频收益的SKST-RS-模型的VaR预测能力强于SKST-RV模型和基于日间收益率的GARCH类模型;在VaR估计市场极端风险失效时,ES能够有效地对尾部极端风险进行管理。
Considering the asymmetric structure of the returns and volatility of China’s stock market index, the partial t distribution is used to fit the biased distribution of returns, and the leverage effects of volatility are captured by RS-ARFIMA-GARCH and SKST-RS-models Respectively, RS- and portrayed the dynamic structure of the yield volatility, thereby improving the VaR and ES and measuring the downside risk of short selling market. Through Kupiec LR and dynamic quantile test, the risk management effects of ES and VaR are empirically analyzed. The results show that SKT-RS-model based on intra-day high-frequency returns has stronger VaR forecasting ability than SKST-RV model and daytime yield-based GARCH model. When VaR estimates extreme risk failure in market, ES can effectively tailor the tail Extreme risk management.