首页 | 本学科首页   官方微博 | 高级检索  
     

基于小波的非参数回归模型多变点检测
引用本文:赵文芝,夏志明,贺兴时. 基于小波的非参数回归模型多变点检测[J]. 西北纺织工学院学报, 2012, 0(4): 536-539
作者姓名:赵文芝  夏志明  贺兴时
作者单位:[1]西安工程大学理学院,陕西西安710048 [2]西北大学数学系,陕西西安710069
基金项目:陕西省教育厅基金资助项目(2010JK561,11JK0188,12JK0744); 西安工程大学基础研究基金资助项目(2010JC07); 国家自然科学基金专项基金项目(数学天元基金)(11026135,11126312); 统计学“青年基金项目”(10YJC910007); 陕西省科技厅软科学项目(2009krm021)
摘    要:构造了基于小波系数的多变点检测算法,该算法可应用于非参数回归模型多变点检测.渚河水文时间序列实例分析结果表明,检测的结果与实际情况相符合.该方法不需要对时间序列做任何参数化的假定便能方便地检测出变点的位置,同时还能够给出变点个数的估计.

关 键 词:多变点  小波系数  小波检测算法  非参数时间序列

Multiple change points detection of nonparametric regression model based on wavelet
ZHAO Wen-zhi,XIA Zhi-ming,HE Xing-shi. Multiple change points detection of nonparametric regression model based on wavelet[J]. Journal of Northwest Institute of Textile Science and Technology, 2012, 0(4): 536-539
Authors:ZHAO Wen-zhi  XIA Zhi-ming  HE Xing-shi
Affiliation:1.School of Science,Xi′an Polytechnic University,Xi′an 710048,China; 2.Department of Mathematics,Northwest University,Xi′an 710069,China)
Abstract:A change point detection algorithm based on the wavelet coefficients is designed to study multiple change points of nonparametric regression model.It is applied to detect change points of the monthly discharge series of Chu river at Hongchun hydrological station,the results of which is consistent with the facts in reality.Moreover,this algorithm can be conveniently used to detect the location of the change point without having to specify a parametric class of time series.Apart from that,the number of change points can be given at the same time.
Keywords:multiple change points  wavelet coefficient  wavelet detection algorithm  nonparametric regression
本文献已被 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号