A note on the variance of a lightly trimmed mean when multiple outliers are present in the sample

被引：3
|
作者：
Balakrishnan, N
David, HA
机构：
[1] McMaster Univ, Dept Math & Stat, Hamilton, ON L8S 4K1, Canada
[2] Iowa State Univ, Dept Stat, Ames, IA 50011 USA
来源：
STATISTICS & PROBABILITY LETTERS | 2001年 / 55卷 / 04期
关键词：
order statistics; product moments; recurrence relation; trimmed mean; L-estimators; location-outliers; scale-outliers; bias; mean square error;
D O I：
10.1016/S0167-7152(00)00238-8
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, a recurrence relation for product moments of order statistics arising from independent and non-identically distributed variables is utilized to derive a convenient formula for the variance of a lightly trimmed mean when one or more outliers are present in the sample. Formulas are presented for the two cases when: (i) the outliers are scale-outliers, and (ii) the outliers are location-outliers. Two examples are given to illustrate the usefulness of the formulas derived in this note. (C) 2001 Elsevier Science B.V. All rights reserved.
引用
下载
收藏
页码：339 / 343
页数：5
相关论文
共 15 条
12