EMPIRICAL LIKELIHOOD RATIO TEST FOR SERIAL CORRELATION IN ADDITIVE MODELS

被引:0
|
作者
Wei, Chuanhua [1 ]
机构
[1] Minzu Univ China, Sch Sci, Dept Stat, Beijing 100081, Peoples R China
基金
中国国家自然科学基金;
关键词
additive models; backfitting; empirical likelihood; serial correlation;
D O I
10.17654/ADASJun2015_153_165
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
We propose an empirical likelihood based method for testing serial correlation in additive models. The proposed test statistic is shown to have asymptotic chi-squared distribution under the null hypothesis of no serial correlation. Furthermore, a simulation study is conducted to illustrate the performance of the proposed method.
引用
收藏
页码:153 / 165
页数:13
相关论文
共 50 条