Some probability inequalities of least-squares estimator in non linear regression model with strong mixing errors

被引:5
|
作者
Yang, Wenzhi [1 ]
Wang, Yiwei [2 ]
Hu, Shuhe [1 ]
机构
[1] Anhui Univ, Sch Math Sci, Hefei, Anhui, Peoples R China
[2] Chinese Univ Hong Kong, Dept Syst Engn & Engn Management, Shatin, Hong Kong, Peoples R China
基金
中国国家自然科学基金;
关键词
Least-squares estimator; Non linear regression model; Strong mixing sequence; LARGE DEVIATION RESULT; SAMPLE QUANTILES; BAHADUR REPRESENTATION; CONVERGENCE; CONSISTENCY; SEQUENCES;
D O I
10.1080/03610926.2014.988261
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
In this paper, we investigate the non linear regression model when the errors are strong mixing. Some probabilityinequalities of the least-squares estimator are presented by using moment information of errors. Meanwhile, for some p > 2, two examples are given when errors satisfy sup(n >= 1) E vertical bar xi(n)vertical bar(p) = infinity and sup(n >= 1) E vertical bar xi(n)vertical bar(p) = infinity, respectively. Then, the complete consistency of the least- squares estimator is obtained.
引用
收藏
页码:165 / 175
页数:11
相关论文
共 50 条