摘要
由于厚尾分布,通常基于正态分布假定的VaR参数法在99%置信水平下计量市场风险时会严重低估风险,文中提出了一种半参数方法。通过对石油市场收益数据的检验,在99%置信水平下,半参数方法的风险计量效果好于通常的参数方法。
Because of the fat tail distribution, the market risk will be underestimated if using the usual parametric model, which is based on the normal distribution hypothesis, to compute the VaR under 99% likelihood level. This paper puts forward a semi-parametric approach to solve this problem, it is tested by the oil market return data. The empirical result shows that it is much better than the usual parametric model under the 99% likelihood level.
出处
《系统工程理论方法应用》
北大核心
2005年第4期379-381,共3页
Systems Engineering Theory·Methodology·Applications
关键词
石油市场收益
受险价值
半参数法
oil market return
Value at Risk(VaR)
semi-parametric approach