摘要
运用Copula模型研究随机变量间的相关结构,是近年来金融统计分析中的一个热点.在龚金国和史代敏提出时变Copula非参数模型的基础上,利用时间序列的极限理论研究了时变参数估计量的大样本性质,并给出了时变Copula模型的非参数估计算法.研究结果表明,时变Copula非参数模型的时变参数估计量具有一致性和渐近正态性.
It becomes a hot topic in the financial statistics analysis that Copula model is applied to discuss the dependence structure among random variables.Based on the time-varying Copula nonparametric model given by GONG Jin-guo and SHI Dai-min(2011),this paper mainly discusses the large sample properties of the estimators of time-varying parameters by the asymptotic theory for time series and gives the algorithm of nonparametric estimation in time-varying Copula model.The result shows that the nonparametric estimators of time-varying parameters have consistency and asymptotic normality.
出处
《浙江大学学报(理学版)》
CAS
CSCD
2012年第6期630-634,642,共6页
Journal of Zhejiang University(Science Edition)
基金
国家社会科学基金青年项目(12CTJ007)
国家自然科学基金面上项目(71171166
71171168)
教育部人文社会科学研究西部和边疆地区项目(11XJC910001)
西南财经大学科研基金资助项目(2011XG119)
西南财大"211工程"三期青年成长项目(211QN2011031)
关键词
时变COPULA
局部极大似然估计
一致性
渐近正态性
time-varying Copula
local maximum likelihood estimation
consistency
asymptotic normality