Fast Computation of Tukey Trimmed Regions and Median in Dimension p > 2

被引:19
|
作者
Liu, Xiaohui [1 ]
Mosler, Karl [2 ]
Mozharovskyi, Pavlo [3 ]
机构
[1] Jiangxi Univ Finance & Econ, Res Ctr Appl Stat, Sch Stat, Nanchang, Jiangxi, Peoples R China
[2] Univ Cologne, Inst Econometr & Stat, Cologne, Germany
[3] Univ Paris Saclay, Telecom ParisTech, LTCI, F-75013 Paris, France
基金
中国博士后科学基金; 中国国家自然科学基金;
关键词
Computational geometry; Depth contours; Depth regions; Halfspace depth; Location depth; R-package TukeyRegion; Tukey depth; Tukey median; REGRESSION DEPTH; LOCATION DEPTH; MULTIVARIATE; CONTOURS;
D O I
10.1080/10618600.2018.1546595
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
Given data in , a Tukey kappa-trimmed region is the set of all points that have at least Tukey depth kappa w.r.t. the data. As they are visual, affine equivariant and robust, Tukey regions are useful tools in nonparametric multivariate analysis. While these regions are easily defined and interpreted, their practical use in applications has been impeded so far by the lack of efficient computational procedures in dimension p > 2. We construct two novel algorithms to compute a Tukey kappa-trimmed region, a naive one and a more sophisticated one that is much faster than known algorithms. Further, a strict bound on the number of facets of a Tukey region is derived. In a large simulation study the novel fast algorithm is compared with the naive one, which is slower and by construction exact, yielding in every case the same correct results. Finally, the approach is extended to an algorithm that calculates the innermost Tukey region and its barycenter, the Tukey median. for this article are available online.
引用
收藏
页码:682 / 697
页数:16
相关论文
共 50 条
  • [31] FAST COMPUTATION OF THE 2-DIMENSIONAL GENERALIZED HARTLEY TRANSFORMS
    HU, NC
    LU, FF
    [J]. IEE PROCEEDINGS-VISION IMAGE AND SIGNAL PROCESSING, 1995, 142 (01): : 35 - 39
  • [32] Fast Genetic Algorithm with Greedy Heuristic for p-Median and k-Means Problems
    Kazakovtsev, Lev A.
    Stupina, Aljona A.
    [J]. 2014 6TH INTERNATIONAL CONGRESS ON ULTRA MODERN TELECOMMUNICATIONS AND CONTROL SYSTEMS AND WORKSHOPS (ICUMT), 2014, : 602 - 606
  • [33] Quick Computation Program of Fractal Dimension for 2D Vector Data
    Wang, Qian
    Wang, Quanfang
    Mei, Xin
    Zhang, Haiwen
    Sun, Hangzhou
    [J]. 2009 17TH INTERNATIONAL CONFERENCE ON GEOINFORMATICS, VOLS 1 AND 2, 2009, : 1128 - 1131
  • [34] ROBUSTNESS OF 2 COMMON HEURISTICS FOR THE P-MEDIAN PROBLEM
    ROSING, KE
    HILLSMAN, EL
    ROSINGVOGELAAR, H
    [J]. ENVIRONMENT AND PLANNING A, 1979, 11 (04) : 373 - 380
  • [35] Modified Two-Dimension Folding Method for Fast P-Code Acquisition
    Guo, Xinpeng
    Zhao, Hongbo
    [J]. 2016 2ND IEEE INTERNATIONAL CONFERENCE ON COMPUTER AND COMMUNICATIONS (ICCC), 2016, : 1777 - 1780
  • [36] Connected Hopf algebras of dimension p2
    Wang, Xingting
    [J]. JOURNAL OF ALGEBRA, 2013, 391 : 93 - 113
  • [37] A Fast Approximate Method for the Large-scale One-source P-median Problem
    Zhao, Runze
    Xiao, Yiyong
    Luo, Rui
    Zhang, Yue
    Liu, Xiaoyuan
    [J]. 2021 IEEE INTERNATIONAL CONFERENCE ON INDUSTRIAL ENGINEERING AND ENGINEERING MANAGEMENT (IEEE IEEM21), 2021, : 1696 - 1700
  • [38] THE COMPUTATION OF ELEMENTARY-FUNCTIONS IN RADIX 2(P)
    MERRHEIM, X
    [J]. COMPUTING, 1994, 53 (3-4) : 219 - 232
  • [39] Evolutionary P2P Networking That Fuses Evolutionary Computation and P2P Networking Together
    Ohnishi, Kei
    Ole, Yuji
    [J]. IEICE TRANSACTIONS ON COMMUNICATIONS, 2010, E93B (02) : 317 - 327
  • [40] A fast algorithm for the computation of 2-D forward and inverse MDCT
    Wu, J. S.
    Shu, H. Z.
    Senhadji, L.
    Luo, L. M.
    [J]. SIGNAL PROCESSING, 2008, 88 (06) : 1436 - 1446