Averaged estimation of functional-coefficient regression models with different smoothing variables

被引：9
|
作者：
Zhang, Riquan [1 ]
Li, Guoying
机构：
[1] E China Normal Univ, Dept Stat, Shanghai 200062, Peoples R China
[2] Shanxi Datong Univ, Dept Math, Datong 037009, Shanxi, Peoples R China
[3] Chinese Acad Sci, Acad Math & Syst Sci, Beijing 100080, Peoples R China
来源：
STATISTICS & PROBABILITY LETTERS | 2007年 / 77卷 / 04期
关键词：
asymptotic normality; different smoothing variables; functional-coefficient regression models; averaged estimate; local linear method;
D O I：
10.1016/j.spl.2006.08.023
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：
The functional-coefficient regression models with different smoothing variables in different coefficient functions are discussed in this paper. The averaged estimates of coefficient functions are defined by averaging the sample on the initial value obtained by a local linear technique. Their asymptotic normality is studied. The efficiency of the proposed method is shown by a simulated example. (c) 2006 Elsevier B.V. All rights reserved.
引用
收藏
页码：455 / 461
页数：7
相关论文
共 50 条
12345