摘要
利用一个半参数门限随机波动率模型来刻画资产收益和波动率之间的关系,该模型可以同时描述资产收益均值、波动率、杠杆效应三种非对称性和未知的收益分布。使用伯努利随机变量来描述非对称性结构,基于非参数密度方法来估计未知分布,利用有效性重要性抽样和惩罚密度方法来估计模型中的未知参数。蒙特卡洛模拟和实例分析说明,该方法在模型参数估计方面有着良好的有限样本表现,并具有实际意义。
This paper proposed a semiparametric threshold stochastic volatility model to describe the relationship between asset return and its volatility.This model could simultaneously capture three types of the asymmetries of average return on assets,volatility and leverage effect,which could also describe unknown distribution of asset return.The asymmetry in this paper was described by Bernoulli random variable.Unknown distribution was estimated by non-parametric density estimation approach.The method based on the efficient important sampling(EIS)and penalized density was taken into account to estimate the model parameters.Monte Carlo simulations and implementation on empirical studies were carried out to assess the finite sample performance of the proposed method in estimating the proposed semiparametric threshold stochastic volatility model,which illustrated the validity of the proposed method in practice.
作者
郝红霞
帅承露
HAO Hong-xia;SHUAI Cheng-lu(School of Statistics and Mathematics,Nanjing Audit University,Nanjing 211815,China)
出处
《统计学报》
2020年第5期46-60,共15页
Journal of Statistics
基金
国家社会科学基金项目(17CTJ016)。
关键词
随机波动率模型
门限效应
非参数建模
有效性重要性抽样
惩罚似然
stochastic volatility model
threshold effect
nonparametric modeling
efficient important sampling
penalty likelihood
