Backtesting Risk Models for the Chinese Metal Futures Market
以中国金属期货市场三个代表性指数为样本，预测八种风险模型的VaR和ES，并通过回测方法评估其准确性，发现带偏斜学生t分布的APARCH模型在描述效率和极端风险估计精度上表现较好。
Take three representative indices of the Chinese metal futures(CMF) market 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 APARCH model with skewed student-t distribution is moderately good in overall considering of description efficiency and estimation accuracy to extreme risk.