摘要
本文提出了T分布的带杠杆效应的随机波动模型,该模型同时兼顾了股票市场的杠杆效应和厚尾效应,并对模型进行了统计结构分析,证明了模型的有效性,基于贝叶斯分析,给出了对ASV-T模型的MCMC估计方法,其中对参数采取Gibbs抽样。利用该模型,通过对中国创业板指数的实证研究,证明了ASV-T模型对创业板市场的回报和波动性特征有更好的拟合效果,并且模型能够较好地描述金融数据的杠杆效应和厚尾效应。
The article empirically study the characteristics of the GEI returns, we find that the returns series have the leverage and fat-tail effect. However, according to the prior research, because there are correlations between the yields, it is difficult in modeling the leverage and fat-tail effect in one model together. The article proposed a new stochastic volatility model considering fat-tail (T-distribution) and leverage effect, it is called ASV-T model. The article made Bayesian statistical distribution parameter estimation and designed the procedure of the Gibbs sampling. According to the empirically study, it is proved that the imitative effect of the ASV-T model in the GEM market fits better than the ASV-N model and the SV-T model.
出处
《数理统计与管理》
CSSCI
北大核心
2016年第6期1086-1097,共12页
Journal of Applied Statistics and Management