Generalized and robustified empirical depths for multivariate data

被引：2
|
作者：
Liu, Xiaohui [1 ,2 ]
Rahman, Jafer [1 ,3 ]
Luo, Shihua [1 ,2 ]
机构：
[1] Jiangxi Univ Finance & Econ, Sch Stat, Nanchang 330013, Jiangxi, Peoples R China
[2] Jiangxi Univ Finance & Econ, Res Ctr Appl Stat, Nanchang 330013, Jiangxi, Peoples R China
[3] Hazara Univ, Dept Math & Stat, Mansehra 21300, Kpk, Pakistan
来源：
STATISTICS & PROBABILITY LETTERS | 2019年 / 146卷
基金：
中国博士后科学基金;
关键词：
General empirical depth; Zonoid depth; Weighted zonoid depth; Weighted mean; PROJECTION DEPTH; CLASSIFICATION;
D O I：
10.1016/j.spl.2018.10.018
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 paper, we point out a common feature of zonoid and empirical depths that gives rise to a novel class of data depths, and provide a possible robustification of their deepest points. This general approach may be useful as it allows the robustification of the zonoid depth and other depths that are centered at the usual mean. Illustrations are also provided to demonstrate the depth contours induced from these depths based on a real data set. (C) 2018 Elsevier B.V. All rights reserved.
引用
收藏
页码：70 / 79
页数：10
相关论文
共 50 条
12345