基于KL分位数估计方法的VaR估计及实证分析  被引量:1

On VaR Estimation and Empirical Analysis Based on KL Quantile Method

在线阅读下载全文

作  者:江伟[1,2] 苏玉华[2,3] 

机构地区:[1]广西贺州学院人事处,广西贺州542899 [2]广西贺州学院符号计算与工程数据处理重点实验室,广西贺州542899 [3]广西贺州学院理学院,广西贺州542899

出  处:《西南师范大学学报(自然科学版)》2016年第1期105-110,共6页Journal of Southwest China Normal University(Natural Science Edition)

基  金:广西教育厅项目(2013LX114);贺州学院科研项目(2014ZC12)

摘  要:为了有效地度量小概率水平下金融产品的风险价值量VaR,在概率水平小于等于0.05的情况下对KL分位数估计进行数值模拟分析,在此基础上提出利用KL分位数估计量去估计VaR的方法.为此先介绍了KL分位数估计子样本容量的选择和均方误差的计算,再分别对KL分位数估计与SQ样本分位数估计做模拟分析并讨论其拟合效果,比较不同分布和不同样本取值下两者的均方误差比率,模拟分析结果表明:一般情况下KL分位数估计优于SQ样本分位数估计.最后,将KL分位数估计分别运用于桂林旅游、桂林三金这两支股票的风险度量(VaR估计),得出投资桂林旅游股票的风险大于投资桂林三金股票的风险.In order to measure the VaR effectively,we have given the KL quantile method to estimate the VaR,which is by doing numerical analysis for the KL quantile estimation with the equal or less than to 0.05 probability leve.For this,we firstly introduce the methods of choosing the subsample size and computing the mean square error of the KL quantile estimating,then for the KL quantile estimation and SQ sample quantile estimation,we do simulation analysis and discusses the fitting effect,and more compare their mean square error ratios under the conditions of different distributions and different sample values,and the simulation shows that the KL quantile estimating is generally superior to the SQ sample quantile one.Furthermore,the KL quantile method is applied to the VaR estimations of two stocks of Guilin tourism and Guilin Sanjin,it is concluded that the risk for the former is bigger than the one for the latter.

关 键 词:KL分位数估计 VAR估计 SQ样本分位数估计 

分 类 号:O212[理学—概率论与数理统计]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象