半参数门限广义非对称随机波动模型研究  

Research on Semiparametric Threshold Generalized Asymmetric Stochastic Volatility Model

作　　者：梁秋阳 陈家清[1] 王仁祥[2] LIANG Qiu-yang;CHEN Jia-qing;WANG Ren-xiang(College of Science,Wuhan University of Technology,Wuhan 430070,China;College of Economics,Wuhan University of Technology,Wuhan 430070,China)

机构地区：[1]武汉理工大学理学院,湖北武汉430070 [2]武汉理工大学经济学院,湖北武汉430070

出　　处：《数学的实践与认识》2021年第5期36-44,共9页Mathematics in Practice and Theory

基　　金：国家自然科学基金面上资助项目(81671633)。

摘　　要：为了能够同时刻画和描述金融资产收益序列的偏态、厚尾以及序列的门限效应、非对称杠杆效应等特性,提出把门限广义非对称随机波动模型与非参数Dirichlet过程混合模型有机结合,构建了半参数门限广义非对称随机波动模型,并对模型进行了贝叶斯分析.实证研究中,利用上海黄金价格收益率序列数据进行建模分析,结果表明:半参数门限广义非对称随机波动模型能够有效地刻画上海黄金价格收益率序列波动率的动态特征.In order to be able to characterize and describe the skewness,thick tail,and the threshold effect and asymmetric leverage effect of the sequence of financial asset returns simultaneously,this paper proposes to combine the threshold generalized asymmetric stochastic volatility model and the nonparametric Dirichlet process hybrid model to construct a semiparametric threshold generalized asymmetric stochastic volatility model,and Bayesian analysis is used to analyze the model.In the empirical study,the Shanghai gold price return sequence data was used for modeling analysis.The empirical analysis result shows that the semiparametric threshold generalized asymmetric stochastic volatility model can effectively characterize the volatility of Shanghai gold price and return series.

关 键 词：TGASV-DPM模型 上海黄金价格 贝叶斯分析 波动率 

分 类 号：F224[经济管理—国民经济] F832.54