Distance-based depths for directional data

DOI10.1002/CJS.11479zbMATH Open1492.62095arXiv1710.00080OpenAlexW2963853078WikidataQ128675112 ScholiaQ128675112MaRDI QIDQ4960935FDOQ4960935

Authors: Giuseppe Pandolfo, Davy Paindaveine, Giovanni C. Porzio 

Publication date: 24 April 2020

Published in: The Canadian Journal of Statistics (Search for Journal in Brave)

Abstract: Directional data are constrained to lie on the unit sphere of~$mathbb{R}^q$ for some~$qgeq2$. To address the lack of a natural ordering for such data, depth functions have been defined on spheres. However, the depths available either lack flexibility or are so computationally expensive that they can only be used for very small dimensions~$q$. In this work, we improve on this by introducing a class of distance-based depths for directional data. Irrespective of the distance adopted, these depths can easily be computed in high dimensions too. We derive the main structural properties of the proposed depths and study how they depend on the distance used. We discuss the asymptotic and robustness properties of the corresponding deepest points. We show the practical relevance of the proposed depths in two applications, related to (i) spherical location estimation and (ii) supervised classification. For both problems, we show through simulation studies that distance-based depths have strong advantages over their competitors.

Full work available at URL: https://arxiv.org/abs/1710.00080

Recommendations

• Ordering directional data: Concepts of data depth on circles and spheres
• Recent advances in directional statistics
• A new concept of quantiles for directional data and the angular Mahalanobis depth
• Connecting pairwise geodesic spheres by depth: DCOPS
• Dirichlet depths for point process

zbMATH Keywords

supervised classificationhyperspheresstatistical deptharc distance depthspherical locationchord distance depthcosine distance depth

Mathematics Subject Classification ID

Asymptotic properties of nonparametric inference (62G20) Directional data; spatial statistics (62H11)

Cited In (16)

• Clustering directional data through depth functions
• The GLD-plot: a depth-based graphical tool to investigate unimodality of directional data
• Robust mean-variance portfolio through the weighted \(L^p\) depth function
• Distance-based directional depth classifiers: a robustness study
• Title not available (Why is that?)
• Multivariate process control charts based on the \(L^p\) depth
• A new concept of quantiles for directional data and the angular Mahalanobis depth
• Recent advances in directional statistics
• Depth and outliers for samples of sets and random sets distributions
• Tukey’s Depth for Object Data
• Spatial quantiles on the hypersphere
• On the optimality of the max-depth and max-rank classifiers for spherical data.
• Theory of angular depth for classification of directional data
• Ordering directional data: Concepts of data depth on circles and spheres
• Dirichlet depths for point process
• Halfspace depth and floating body

This page was built for publication: Distance-based depths for directional data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4960935)