On an asymptotically more efficient estimation of the single-index model

被引：37

作者：

Chang, Ziqing

Xue, Liugen ^{[2
]}

Zhu, Lixing ^{[1
]}

机构：

[1] Hong Kong Baptist Univ, Dept Math, Hong Kong, Hong Kong, Peoples R China

[2] Beijing Univ Technol, Beijing, Peoples R China

来源：

JOURNAL OF MULTIVARIATE ANALYSIS | 2010年 / 101卷 / 08期

基金：

中国国家自然科学基金;

关键词：

The single-index model; Asymptotical efficiency; Least squares estimation; SEMIPARAMETRIC ESTIMATION; EMPIRICAL LIKELIHOOD;

D O I：

10.1016/j.jmva.2010.02.005

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 note, we revisit the single-index model with heteroscedastic error, and recommend an estimating equation method in terms of transferring restricted least squares to unrestricted least squares: the estimator of the index parameter is asymptotically more efficient than existing estimators in the literature in the sense that it is of a smaller limiting variance. (C) 2010 Elsevier Inc. All rights reserved.

引用

页码：1898 / 1901

页数：4

共 19 条

[1]

[Anonymous], 1996, Local polynomial modelling and its applications

[2]

Bickel P. J., 1993, Efficient and Adaptive Inference in Semiparametric Models

[3]

Brillinger D. R., 1983, A Festschrift for Erich L. Lehmann in Honor of His Sixty-Fifth Birthday, P97

[4] Generalized partially linear single-index models [J].