纺织高校基础科学学报2016，Vol.29Issue(3)：351-355,5.DOI:10.13338/j.issn.1006-8341.2016.03.013
非参数回归模型变点两步估计法及实证分析
Two-step change point estimation in nonparametric regression model and the empirical analysis
摘要
Abstract
The two-step estimators for change point in nonparametric regression are proposed.In the first step,an initial estimator is obtained by local linear smoothing method.In the second step,the fi-nal estimator is obtained by CUSUM method on a closed neighborhood of initial estimator.It is found through a simulation study that the proposed estimator is efficient.The estimator for j ump size is also obtained.Further more,experimental results that using historical data on Nile river discharges,ex-change rate data of USD against RMB and global temperature data for the northern hemisphere show that the proposed method is also practical in applications.关键词
非参数回归模型/变点/局部线性方法/CUSUMKey words
nonparametric regression model/change point/local linear method/CUSUM分类
数理科学引用本文复制引用
赵文芝,夏志明,贺飞跃..非参数回归模型变点两步估计法及实证分析[J].纺织高校基础科学学报,2016,29(3):351-355,5.基金项目
国家自然科学青年基金资助项目(11201372) （11201372）
陕西省教育厅科研计划项目(2013JK0592) （2013JK0592）
