Local Polynomial Estimation of Time-Dependent Diffusion Parameter for Discretely Observed SDE Models

被引:1
|
作者
Wang, Ji-xia [1 ]
Xiao, Qing-xian [2 ]
机构
[1] Henan Normal Univ, Coll Math & Informat Sci, Xinxiang 453007, Peoples R China
[2] Shanghai Univ Sci & Technol, Sch Business, Shanghai 200093, Peoples R China
基金
中国国家自然科学基金;
关键词
Diffusion model; Time-dependent parameter; Local polynomial estimation; Kernel weighted; Asymptotic normality; TERM STRUCTURE DYNAMICS;
D O I
10.2298/FIL1404871W
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
Extending the results of Yu, Yu, Wang and Lin [10], we study the local polynomial estimation of the time-dependent diffusion parameter for time-inhomogeneous diffusion models. Considering the diffusion parameter being positive, we obtain the local polynomial estimation of the diffusion parameter by taking the diffusion parameter to be local log-polynomial fitting. The asymptotic bias, asymptotic variance and asymptotic normal distribution of the volatility function are discussed. A real data analysis is conducted to show the performance of the estimations proposed.
引用
收藏
页码:871 / 878
页数:8
相关论文
共 50 条