Some intriguing properties of Tukey's half-space depth

DOI10.3150/10-BEJ322zbMATH Open1229.62063arXiv1201.1171OpenAlexW2016312261MaRDI QIDQ654416FDOQ654416

Authors: Subhajit Dutta, Probal Chaudhuri, Anil K. Ghosh 

Publication date: 28 December 2011

Published in: Bernoulli (Search for Journal in Brave)

Abstract: For multivariate data, Tukey's half-space depth is one of the most popular depth functions available in the literature. It is conceptually simple and satisfies several desirable properties of depth functions. The Tukey median, the multivariate median associated with the half-space depth, is also a well-known measure of center for multivariate data with several interesting properties. In this article, we derive and investigate some interesting properties of half-space depth and its associated multivariate median. These properties, some of which are counterintuitive, have important statistical consequences in multivariate analysis. We also investigate a natural extension of Tukey's half-space depth and the related median for probability distributions on any Banach space (which may be finite- or infinite-dimensional) and prove some results that demonstrate anomalous behavior of half-space depth in infinite-dimensional spaces.

Full work available at URL: https://arxiv.org/abs/1201.1171

Recommendations

• The depth function of a population distribution.
• General notions of statistical depth function.
• On the behavior of Tukey's depth and median under symmetric stable distributions.
• An algorithm for the multivariate Tukey median
• Some results on the computing of Tukey's halfspace median

zbMATH Keywords

symmetric distributionsBanach space\(l_p\) normdepth contourshalf-space median

Mathematics Subject Classification ID

Characterization and structure theory for multivariate probability distributions; copulas (62H05) Applications of functional analysis in probability theory and statistics (46N30) Graphical methods in statistics (62A09)

Cites Work

• Breakdown properties of location estimates based on halfspace depth and projected outlyingness
• The multivariate L ₁ -median and associated data depth
• Multivariate analysis by data depth: Descriptive statistics, graphics and inference. (With discussions and rejoinder)
• On data depth and distribution-free discriminant analysis using separating surfaces
• General notions of statistical depth function.
• On Maximum Depth and Related Classifiers
• Multivariate dispersion, central regions and depth. The lift zonoid approach
• On a notion of data depth based on random simplices
• Projection-based depth functions and associated medians
• Title not available (Why is that?)
• Location–Scale Depth
• Sign Tests in Multidimension: Inference Based on the Geometry of the Data Cloud
• Title not available (Why is that?)
• Asymptotics for multivariate trimming
• Nonparametric depth-based multivariate outlier identifiers, and masking robustness properties
• The Tukey and the random Tukey depths characterize discrete distributions
• The Tukey depth characterizes the atomic measure
• Title not available (Why is that?)
• A simple test for uniformity of a circular distribution
• On the Tukey depth of a continuous probability distribution

Cited In (36)

• Statistical depth in abstract metric spaces
• Common multivariate estimators of location and scatter capture the symmetry of the underlying distribution
• Data depth for measurable noisy random functions
• The depth function of a population distribution.
• Finite sample breakdown point of Tukey's halfspace median
• On a general definition of depth for functional data
• A topologically valid definition of depth for functional data
• Concentration of the empirical level sets of Tukey's halfspace depth
• Depth in Infinite-dimensional Spaces
• On the behavior of Tukey's depth and median under symmetric stable distributions.
• A centrality notion for graphs based on Tukey depth
• Weighted lens depth: Some applications to supervised classification
• Approximate calculation of Tukey's depth and median with high-dimensional data
• Integrated depth for functional data: statistical properties and consistency
• Halfspace depth does not characterize probability distributions
• Depth and outliers for samples of sets and random sets distributions
• On the computation of multivariate scenario sets for the skew-\(t\) and generalized hyperbolic families
• Depth-weighted Bayes classification
• Choosing among notions of multivariate depth statistics
• Half-region depth for stochastic processes
• Convergence of quantile and depth regions
• From Depth to Local Depth: A Focus on Centrality
• A Bayesian Approach to Multiple-Output Quantile Regression
• Tukey’s Depth for Object Data
• A weighted localization of halfspace depth and its properties
• A Decomposition of Total Variation Depth for Understanding Functional Outliers
• Depth for curve data and applications
• On smoothness of Tukey depth contours
• On the Tukey depth of a continuous probability distribution
• Model-based statistical depth with applications to functional data
• Half-space mass: a maximally robust and efficient data depth method
• Halfspace depth and floating body
• A topologically valid construction of depth for functional data
• On data depth in infinite dimensional spaces
• On weighted and locally polynomial directional quantile regression
• Halfspace depth and regression depth characterize the empirical distribution

This page was built for publication: Some intriguing properties of Tukey's half-space depth

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q654416)