High-dimensional computation of the deepest location

被引：48

作者：

Struyf, A ^{[1
]}

Rousseeuw, PJ ^{[1
]}

机构：

[1] Univ Instelling Antwerp, Dept Math & Comp Sci, B-2610 Antwerp, Belgium

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 2000年 / 34卷 / 04期

关键词：

algorithm; location depth; multivariate median;

D O I：

10.1016/S0167-9473(99)00112-7

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

The halfspace location depth of a point theta relative to a data set X-n is defined as the smallest number of observations in any closed halfspace with boundary through theta. As such, halfspace depth can be seen as a kind of multivariate ranking. The deepest location, i.e. the theta with maximal halfspace depth, is a multivariate generalization of the median. Until now the deepest location could only be computed for bivariate data. In this paper, we construct an algorithm (called DEEPLOC) to approximate the deepest location in higher dimensions. (C) 2000 Elsevier Science B.V. All rights reserved.

引用

页码：415 / 426

页数：12

共 24 条

[1]

ALFURIAH I, 1998, DIMACS WORKSH EXT ME

[2]

BAI ZD, 1999, IN PRESS ANN STAT, V27

[3] BREAKDOWN PROPERTIES OF LOCATION ESTIMATES BASED ON HALF-SPACE DEPTH AND PROJECTED OUTLYINGNESS [J].

DONOHO, DL ;

GASKO, M .

ANNALS OF STATISTICS, 1992, 20 (04) :1803-1827

[4]

Dyckerhoff R., 1996, COMPSTAT. Proceedings in Computational Statistics. 12th Symposium, P235

[5]

Hand D. J., 1994, HDB SMALL DATA SETS

[6]

Hettmansperger T., 1992, L1 STAT ANAL RELATED, P267

[7]

HETTMANSPERGER TP, 1994, J ROY STAT SOC B MET, V56, P235

[8]

Johnson T., 1998, Proceedings Fourth International Conference on Knowledge Discovery and Data Mining, P224

[9] Notions of limiting P values based on data depth and bootstrap [J].

Liu, RY ;

Singh, K .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1997, 92 (437) :266-277

[10]

Liu RY, 1999, ANN STAT, V27, P783

← 1 2 3 →