沪深300股指期货的风险测度模型研究

Backtesting Risk Models for CSI300 Index Futures

以沪深300股指期货指数的30分钟交易数据为例,首先对其价格变化的动力学特征及波动模式进行了全面深入的考察,然后运用严谨系统的后验分析(Backtesting analysis)方法,分别在多头和空头两种头寸状况以及5种不同分位数水平下,实证对比了8种风险测度模型对VaR(Value at Risk)和ES(Excepted shortfall)两种不同风险指标估计的精度差异。研究结果表明:我国股指期货市场的价格波动具有较为明显的有偏和尖峰厚尾分布、聚集特征和长记忆性;采用有偏学生t分布和长记忆模型有助于提高对沪深300股指期货的风险测度精度,而在波动模型中包含杠杆效应项对提高风险估计精度并无太多帮助;在综合考虑了模型对沪深300股指期货价格变化动力学的刻画效果以及对不同风险指标的测度精度等因素后,基于有偏学生t分布的GARCH模型是一个相对合理的风险测度模型选择。

Take 30-minutes data of CSI300 index as sample, this paper carries out VaR and ES predicting for eight risk models. Furthermore, backtesting methodologies are introduced to estimate the accuracy for VaR and ES predictions produced by different models. The main results show that there is significant leverage effect and long memory in price volatility of CSI300 and clear skewness and fat-tail are observed. Skewed student-t distribution is helpful to improve results in estimating VaR and ES of CSI300. In addition, GARCH model with skewed student-t distribution is moderately good in overall considering of description efficiency and estimation accuracy to extreme risk.