摘要: |
利用WTI日对数收益率数据估计一般极值分布参数,并根据极值BMM模型计算石油价格极端值风险。实证结果表明,在95%的置信水平下,由BMM模型度量的石油价格风险低于正态模型的相应度量值,而在99%的置信水平下,BMM模型对风险的捕捉则显著优于正态分布。上述结论表明,BMM模型在较低置信水平其效力虽不及正态分布VaR模型,但在高置信水平下能更好地捕捉分布的厚尾特征。因此,对于石油风险管理者而言,BMM模型将是测度石油市场极端风险较好的选择。 |
关键词: 极值理论 BMM模型 石油价格 风险度量 |
DOI:10.13216/j.cnki.upcjess.2015.04.0002 |
分类号:F810 |
基金项目:国家自然科学基金面上项目(71071132) |
|
Risk Measurement of Oil Price by Implementing BMM Model |
LIU Fei1, ZHENG Xiaoya2
|
(1.Guangzhou Rural Commercial Bank, Guangzhou, Guangdong 510623, China;2.Post-doctoral Research Station, Shanghai University of Finance and Economics, Shanghai 200433, China)
|
Abstract: |
The paper estimates parameters on the general extreme value distribution using WTI return and we calculate the risk of oil price using VaR formula deduced from the BMM model. The results show that the risk measured by BMM model is lower than by the normal model in the 95% confidence level, and vice versa, in the 99% confidence level. It illustrates that the higher confidence level is the more BMM model which can capture the heavy tail characteristics of the distribution. However, the effectiveness is less than normal distribution in the lower confidence level. Therefore, BMM model will be a better choice to measure the extreme risk for manager in oil market. |
Key words: extreme value theory BMM model oil price risk measurement |