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
  • [1] Fast Computation of Trimmed Means
    Beliakov, Gleb
    [J]. JOURNAL OF STATISTICAL SOFTWARE, 2011, 39 (CS2):
  • [2] On Exact Computation of Tukey Depth Central Regions
    Fojtik, Vit
    Laketa, Petra
    Mozharovskyi, Pavlo
    Nagy, Stanislav
    [J]. JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS, 2024, 33 (02) : 699 - 713
  • [3] Computing zonoid trimmed regions of dimension d > 2
    Mosler, Karl
    Lange, Tatjana
    Bazovkin, Pavel
    [J]. COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2009, 53 (07) : 2500 - 2510
  • [4] An Exact Algorithm for Weighted-Mean Trimmed Regions in Any Dimension
    Bazovkin, Pavel
    Mosler, Karl
    [J]. JOURNAL OF STATISTICAL SOFTWARE, 2012, 47 (13): : 1 - 29
  • [5] FAST COMPUTATION OF OPTIMAL PATHS IN 2-DIMENSION AND HIGHER-DIMENSION MAPS
    HASSOUN, MH
    SANGHVI, AJ
    [J]. NEURAL NETWORKS, 1990, 3 (03) : 355 - 363
  • [6] Partitioning trimmed spline surfaces into nonself-occluding regions for visibility computation
    Krishnan, S
    Manocha, D
    [J]. GRAPHICAL MODELS, 2000, 62 (04) : 283 - 307
  • [7] Fast algorithm for vector median filters with vector distance computation
    Zhong, Ling
    Zhang, Yun
    [J]. Huazhong Keji Daxue Xuebao (Ziran Kexue Ban)/Journal of Huazhong University of Science and Technology (Natural Science Edition), 2010, 38 (09): : 31 - 34
  • [8] Fast median computation for symmetric, orthogonal matrices under the rank distance
    Meidanis, Joao
    Chindelevitch, Leonid
    [J]. LINEAR ALGEBRA AND ITS APPLICATIONS, 2021, 614 : 394 - 414
  • [9] Vector-median type filters and fast-computation algorithms
    Lucat, L
    Siohan, P
    [J]. ISCAS '97 - PROCEEDINGS OF 1997 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS, VOLS I - IV: CIRCUITS AND SYSTEMS IN THE INFORMATION AGE, 1997, : 2469 - 2472
  • [10] Fast computation of fractal dimension for 2D, 3D and 4D data
    Ruiz de Miras, J.
    Posadas, M. A.
    Ibanez-Molina, A. J.
    Soriano, M. F.
    Iglesias-Parro, S.
    [J]. JOURNAL OF COMPUTATIONAL SCIENCE, 2023, 66