摘要
为了更好地刻画金融市场波动率的尖峰厚尾、集聚性、长记忆性、杠杆效应以及非对称等特征,基于上证综合指数和深证成份指数日收益数据,考虑将门限效应和杠杆效应引入分式O-U随机过程,同时考虑杠杆参数是状态相关的,对资产收益波动率进行建模;运用基于贝叶斯原理的马尔科夫链蒙特卡罗(MCMC)方法估计模型中的参数并进行实证分析。研究结果表明,中国股票市场资产价格具有较强的波动率持续性、较高的波动性水平和杠杆效应,利好消息和利空消息对中国股票市场的影响具有非对称性。
In order to better describe the characteristics of financial market volatility,such as sharp peak and thick tail,clustering,long memory,leverage and asymmetry,both the threshold effect and leverage effect are incorporated into the fractional O-U stochastic process in this paper,which is based on daily returns data of Shanghai composite index and Shenzhen component index.At the same time,it is proposed to model the volatility of asset returns considering the leverage parameter is state-correlative.The authors use MCMC method which is based on Bayesian theory to estimate the parameters in this model and then analyze it empirically.The results show that Chinese stock markets have many characteristics,such as stronger volatility persistence,higher volatility level and leverage effect.Besides,good news and bad news have asymmetrical effect on China stock markets.
作者
毛小丽
王仁曾
MAO Xiao-li;WANG Ren-zeng(School of Economics and Commerce,South China University of Technology,Guangzhou 510006,China)
出处
《兰州财经大学学报》
2018年第6期23-32,共10页
Journal of Lanzhou University of Finance and Economics
基金
广州市哲学社会科学规划智库课题(2016GZZK29)
关键词
分式布朗运动
随机波动率
门限效应
杠杆效应
MCMC模拟
Fractional Brownian Motion
stochastic volatility
threshold effect
leverage effect
MCMC simulation
