THE LAWS OF THE ITERATED LOGARITHM OF SOME ESTIMATES IN PARTLY LINEAR-MODELS

被引：36

作者：

GAO, JT ^{[1
]}

机构：

[1] UNIV AUCKLAND,DEPT STAT,AUCKLAND,NEW ZEALAND

来源：

STATISTICS & PROBABILITY LETTERS | 1995年 / 25卷 / 02期

基金：

中国国家自然科学基金;

关键词：

THE LAW OF ITERATED LOGARITHM; LEAST-SQUARE ESTIMATOR; PARTLY LINEAR MODEL;

D O I：

10.1016/0167-7152(94)00217-V

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：

Consider the regression model Y-i = x'(i) beta + g(t(i)) + e(i), 1 less than or equal to i less than or equal to n, where x(i) = (x(i1),x(i2),..., x(ip))' and t(i)(t(i) is an element of [0,1]) are known and nonrandom design points, beta = (beta(1),...,beta(p))' (p greater than or equal to 1) is an unknown parameter, g(.) is an unknown function, and e(i) are i.i.d. random errors. Based on g estimated by nonparametric kernel estimation, the laws of the iterated logarithm of the least-square estimator of beta and an estimator of sigma(2) = Ee(1)(2) < infinity are investigated.

引用

页码：153 / 162

页数：10

共 15 条

[1]

Ansley CF, 1983, APPLIED TIME SERIES, P181

[2] CONVERGENCE-RATES FOR PARAMETRIC COMPONENTS IN A PARTLY LINEAR-MODEL [J].