非线性模型中VaR估测的正态性改进

VaR风险测度技术已经被学界和业界广泛使用,但其局限性也是显而易见的,国内外学者对其进行了一系列的改进.由线性模型扩展为非线性模型以及由正态假定转换到非正态性均源于风险测度的精确化.探讨依数据特征改进和扩展VaR估测方法,使用Johnson转换方法与Cornish-Fisher扩展方法这两种正态性改进方法改善VaR估值,一方面利用正态假定成熟理论结果简化VaR估测方法的推演,另一方面从实证分析角度论证了正态性改进方法在VaR估测中的准确性与有效性.

Improvement Based on Normality about Estimation of VaR in Nonlinear Model

Risk measure methods-VaR has been widely used in academic circles and industry, but its limitations are obvious,so many scholars at home and abroad have done lot of improve models,to switch assumption of normal distribution to non-normality.The study explores these methods to estimate Var which are improved and expanded according to data feature. This compares between the Delta-normal method and the Delta-Gamma-Johnson transition method,between the Delta-normal method and Cornish-Fisher extension method,then proves volidity and veracity by empirical analysis to these extension methods based on normality which is estimated to VaR.