基于Bootstrap的金属期货市场风险VaR区间预测  

Research on Interval Forecast For Metal Futures Market's VaR Based on Bootstrap

作　　者：沈盟[1] 王璐[1,2] 

机构地区：[1]西南交通大学数学学院,四川成都610031 [2]西南交通大学经济管理学院,四川成都610031

出　　处：《武汉理工大学学报（信息与管理工程版）》2016年第1期37-41,共5页Journal of Wuhan University of Technology：Information & Management Engineering

基　　金：国家自然科学基金资助项目(71201131);中国博士后科学基金资助项目(2014M562334)

摘　　要：金属期货市场风险VaR的准确测度对防范期货交易风险及保持市场健康平稳运行有重要作用。传统的VaR测度方法主要以点预测为主,无法反映预测近似值的精确程度及范围。因此,提出了一种基于Bootstrap的金属期货市场风险VaR区间预测方法,同时引入LR检验区间预测的有效性,最后利用我国铜和铝期货市场数据进行了VaR风险的区间预测。结果表明,新的VaR区间预测方法能克服点预测的不足,准确有效地描述VaR的估计风险,同时置信区间上下限可用于风险的预警及控制。The accurate measurement on metal futures market＇s VaR is important to prevent the futures＇ transaction risk and maintain the market healthy and stable operation. The traditional methods measuring VaR focus on point forecast which can＇t reflect the accuracy of the predictive value. A new method to measure the interval forecast for metal futures market＇s VaR based on bootstrap is put forward. LR test is used to test the effectiveness of interval forecast. Finally,we empirical study the interval forecast of copper and aluminum futures market＇s VaR in China. The results show that the new method can overcome the lack of point forecast,while the upper and lower confidence interval can be used for early warning and control financial risks.

关 键 词：金属期货市场 VAR BOOTSTRAP 区间预测 

分 类 号：F224[经济管理—国民经济]