Tukey’s Depth for Object Data
From MaRDI portal
Publication:6077569
Abstract: We develop a novel exploratory tool for non-Euclidean object data based on data depth, extending the celebrated Tukey's depth for Euclidean data. The proposed metric halfspace depth, applicable to data objects in a general metric space, assigns to data points depth values that characterize the centrality of these points with respect to the distribution and provides an interpretable center-outward ranking. Desirable theoretical properties that generalize standard depth properties postulated for Euclidean data are established for the metric halfspace depth. The depth median, defined as the deepest point, is shown to have high robustness as a location descriptor both in theory and in simulation. We propose an efficient algorithm to approximate the metric halfspace depth and illustrate its ability to adapt to the intrinsic data geometry. The metric halfspace depth was applied to an Alzheimer's disease study, revealing group differences in the brain connectivity, modeled as covariance matrices, for subjects in different stages of dementia. Based on phylogenetic trees of 7 pathogenic parasites, our proposed metric halfspace depth was also used to construct a meaningful consensus estimate of the evolutionary history and to identify potential outlier trees.
Recommendations
- Tukey depth histograms
- Approximating Tukey's Depth
- The random Tukey depth
- Some extensions of Tukey's depth function
- Fast implementation of the Tukey depth
- Absolute approximation of Tukey depth: theory and experiments
- Approximate calculation of Tukey's depth and median with high-dimensional data
- Some intriguing properties of Tukey's half-space depth
- On the Tukey depth of a continuous probability distribution
Cites work
- A characterization of halfspace depth
- A new approach for the computation of halfspace depth in high dimensions
- A nonparametric multivariate multisample test based on data depth
- A Quality Index Based on Data Depth and Multivariate Rank Tests
- A smeary central limit theorem for manifolds with application to high-dimensional spheres
- Approximate calculation of Tukey's depth and median with high-dimensional data
- Asymptotics for the Tukey depth process, with an application to a multivariate trimmed mean
- Averages of unlabeled networks: geometric characterization and asymptotic behavior
- Breakdown properties of location estimates based on halfspace depth and projected outlyingness
- Bridging centrality and extremity: refining empirical data depth using extreme value statistics
- Connecting pairwise geodesic spheres by depth: DCOPS
- DD-classifier: nonparametric classification procedure based on DD-plot
- Descriptive statistics for multivariate distributions
- Distance-based depths for directional data
- Exact computation of the halfspace depth
- Finite sample breakdown point of Tukey's halfspace median
- General notions of statistical depth function.
- Geometry of the space of phylogenetic trees
- Globally convergent optimization algorithms on Riemannian manifolds: Uniform framework for unconstrained and constrained optimization
- Halfspace depths for scatter, concentration and shape matrices
- scientific article; zbMATH DE number 3541764 (Why is no real title available?)
- scientific article; zbMATH DE number 3053873 (Why is no real title available?)
- Intrinsic Data Depth for Hermitian Positive Definite Matrices
- Introduction to Riemannian Manifolds
- Large sample properties of the regression depth induced median
- Large sample theory of intrinsic and extrinsic sample means on manifolds. II.
- Measures of centrality for multivariate and directional distributions
- Modeling and Inference for Measured Crystal Orientations and a Tractable Class of Symmetric Distributions for Rotations in Three Dimensions
- Modeling sparse longitudinal data on Riemannian manifolds
- Multidimensional medians arising from geodesics on graphs
- Nonparametric statistics of dynamic networks with distinguishable nodes
- On 64%-Majority Rule
- On a notion of data depth based on random simplices
- On general notions of depth for regression
- On the Concept of Depth for Functional Data
- Overview of object oriented data analysis
- Principal component analysis and the locus of the Fréchet mean in the space of phylogenetic trees
- Principal component analysis for functional data on Riemannian manifolds and spheres
- Regression Depth
- Robust covariance and scatter matrix estimation under Huber's contamination model
- Some special Vapnik-Chervonenkis classes
- Statistical shape analysis. With applications in R
- Statistics on stratified spaces
- Torus principal component analysis with applications to RNA structure
- Trimmed means for functional data
- Weak convergence and empirical processes. With applications to statistics
Cited in
(16)- Fast implementation of the Tukey depth
- Spatial quantiles on the hypersphere
- Metric statistics: exploration and inference for random objects with distance profiles
- Robust inference theory for non-regular time series models and its extensions
- Nonparametric Statistical Inference via Metric Distribution Function in Metric Spaces
- Explicit bivariate simplicial depth
- Regularized halfspace depth for functional data
- Halfspace depth
- Projection depth and L^r-type depths for fuzzy random variables
- Rejoinder to the discussion on ``Exploratory functional data analysis
- Comments on: ``Exploratory functional data analysis
- Wasserstein-Fréchet integration of conditional distributions
- Spatial depth for data in metric spaces
- Nonparametric two-sample test for random objects based on data depth
- Measure of shape for object data
- Weighted embedding and outlier detection of metric space data
This page was built for publication: Tukey’s Depth for Object Data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6077569)
