KERNEL ESTIMATION FOR ADDITIVE-MODELS UNDER DEPENDENCE

被引：7

作者：

BAEK, J ^{[1
]}

WEHRLY, TE ^{[1
]}

机构：

[1] TEXAS A&M UNIV SYST,DEPT STAT,COLL STN,TX 77843

来源：

STOCHASTIC PROCESSES AND THEIR APPLICATIONS | 1993年 / 47卷 / 01期

关键词：

MIXING CONDITIONS; NONPARAMETRIC REGRESSION; OPTIMAL RATE OF CONVERGENCE; TIME SERIES; NADARAYA-WATSON ESTIMATOR;

D O I：

10.1016/0304-4149(93)90096-M

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：

Nonparametric estimation of the conditional mean function for additive models is investigated in cases where the observed data are dependent. We use an additive kernel estimator which is a sum of Nadaraya-Watson estimators. Under a strong mixing condition, the kernel estimator is shown to be asymptotically normal and to achieve the univariate optimal rate of convergence in mean squared error.

引用

页码：95 / 112

页数：18

共 25 条

[1]

BAEK J, 1992, KERNEL REGRESSION ES

[2]

BIERENS HJ, 1983, J AM STAT ASSOC, V78, P699

[3]

Billingsley P, 1968, CONVERGENCE PROBABIL

[4] ASYMPTOTIC-DISTRIBUTION OF ROBUST ESTIMATORS FOR NONPARAMETRIC MODELS FROM MIXING PROCESSES [J].