摘要
传统VaR方法在衡量投资组合的风险上存在诸多缺陷,针对BDSS模型进行改进,基于相对价差得到了再修正的BDSS模型以计量流动性风险.实证分析表明,Copula-kernel模型对多元收益率和相对价差序列的拟合程度都很高,且La-VaR中的流动性风险部分随着置信度的减小而逐渐显著,返回测试表明无论置信度的高低,在大多数情况下La-VaR都不会低估风险.
Traditional VaR method has many defects in measuring portfolio risk,this paper modifies BDSS model based on relative spread to measnre ligmidity risk. The empirical analysis shows that Copula-kernel model can accurately fit multiple yield and relative spread order, and the part of liquidity risk in the La-VaR is gradually significant with the decrease of confidence c. The back testing shows that La-VaR may not underestimate risk in most cases whatever confidence c is high or low.
出处
《河南科学》
2013年第4期547-551,共5页
Henan Science