具有有偏厚尾的非对称SV模型及其实证研究

为了描述资产收益与波动率之间的非对称关系,提出一种非对称SV模型,即具有杠杆效应与尺寸效应的SV(SV-LS)模型。进一步,针对资产收益分布展现出"有偏"及"厚尾"特征,引入有偏广义误差分布(SGED)来描述资产收益,提出具有有偏厚尾的SGED假定下的SV-LS模型。继而,基于有效重要性抽样(EIS)技巧,给出了模型参数的极大似然(ML)估计方法。最后,采用上证综合指数收益数据进行实证研究。结果表明,SGED假定下的SV-LS模型表现最优,它能够综合刻画资产收益的"有偏"及"厚尾"特征,并且证明了我国沪市具有很强的波动持续性以及显著的杠杆效应。

Empirical Study of Asymmetric SV Model with Skewed and Heavy Tailed Distribution

In this paper,we propose a asymmetric SV model,namely the SV model with leverage and size effects(SV-LS),by modeling the asymmetric correlation between return and volatility innovations.We also consider the skewed generalized error distribution(SGED) for capturing skewness and heavy tails.Then,we propose a method for maximum likelihood(ML) estimation of this model based on the efficient importance sampling(EIS) technique.Finally,an empirical study of Shanghai Stock Exchange composite index is presented.Empirical results show that the SV-LS model based on the SGED performs better than other models,which can better account for skewed and heavy-tailed returns.Furthermore,the results also show that in Shanghai stock market there exists high persistence of volatility and a significant leverage effect.