Bridging centrality and extremity: refining empirical data depth using extreme value statistics

DOI10.1214/15-AOS1359MaRDI QIDQ892256zbMATH OpenFDO

Authors John H. J. Einmahl, Jun Li, Regina Y. Liu

Publication date 18 November 2015

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1510.08694, https://projecteuclid.org/euclid.aos/1444222091

zbMATH Keywords

depthextremesnonparametric classificationtailnonparametric multivariate SPC

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Asymptotic properties of nonparametric inference (62G20) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Applications of statistics in engineering and industry; control charts (62P30) Statistics of extreme values; tail inference (62G32)

Abstract: Statistical depth measures the centrality of a point with respect to a given distribution or data cloud. It provides a natural center-outward ordering of multivariate data points and yields a systematic nonparametric multivariate analysis scheme. In particular, the half-space depth is shown to have many desirable properties and broad applicability. However, the empirical half-space depth is zero outside the convex hull of the data. This property has rendered the empirical half-space depth useless outside the data cloud, and limited its utility in applications where the extreme outlying probability mass is the focal point, such as in classification problems and control charts with very small false alarm rates. To address this issue, we apply extreme value statistics to refine the empirical half-space depth in "the tail." This provides an important linkage between data depth, which is useful for inference on centrality, and extreme value statistics, which is useful for inference on extremity. The refined empirical half-space depth can thus extend all its utilities beyond the data cloud, and hence broaden greatly its applicability. The refined estimator is shown to have substantially improved upon the empirical estimator in theory and simulations. The benefit of this improvement is also demonstrated through the applications in classification and statistical process control.

Recommendations

• General notions of statistical depth function.
• Data depth: multivariate statistics and geometry
• Concept of data depth and its applications
• An empirical depth function for multivariate data
• Generalized and robustified empirical depths for multivariate data

Cites work

• scientific article; zbMATH DE number 5604036 (Why is no real title available?)
• scientific article; zbMATH DE number 3541764 (Why is no real title available?)
• scientific article; zbMATH DE number 1086063 (Why is no real title available?)
• scientific article; zbMATH DE number 3023295 (Why is no real title available?)
• A Quality Index Based on Data Depth and Multivariate Rank Tests
• A moment estimator for the index of an extreme-value distribution
• A simple general approach to inference about the tail of a distribution
• Breakdown properties of location estimates based on halfspace depth and projected outlyingness
• Control Charts for Multivariate Processes
• DD-classifier: nonparametric classification procedure based on DD-plot
• Estimation of extreme risk regions under multivariate regular variation
• Extreme value theory. An introduction.
• General notions of statistical depth function.
• Limit distributions for sums of independent random vectors. Heavy tails in theory and practice
• Multivariate analysis by data depth: Descriptive statistics, graphics and inference. (With discussions and rejoinder)
• Multivariate quantiles and multiple-output regression quantiles: from \(L_{1}\) optimization to halfspace depth
• Multivariate spacings based on data depth. I: Construction of nonparametric multivariate tolerance regions
• New nonparametric tests of multivariate locations and scales using data depth
• Notions of Limiting P Values Based on Data Depth and Bootstrap
• On a Geometric Notion of Quantiles for Multivariate Data
• On a notion of data depth based on random simplices
• Projection-based depth functions and associated medians
• Rates of growth and sample moduli for weighted empirical processes indexed by sets
• Regression Depth
• Regularly varying functions
• The convex hull of a random set of points
• The random Tukey depth

Cited in

(12)

• scientific article; zbMATH DE number 7800975 (Why is no real title available?)
• Half-space mass: a maximally robust and efficient data depth method
• Affine invariant integrated rank-weighted statistical depth: properties and finite sample analysis
• Tukey’s Depth for Object Data
• Halfspace depth does not characterize probability distributions
• On multivariate separating Hill estimator under estimated location and scatter
• Illumination Depth
• Halfspace depth and floating body
• Extreme value theory for anomaly detection -- the GPD classifier
• Model-based statistical depth with applications to functional data
• Nonparametric Imputation by Data Depth
• Nonparametric Fusion Learning for Multiparameters: Synthesize Inferences From Diverse Sources Using Data Depth and Confidence Distribution

This page was built for publication: Bridging centrality and extremity: refining empirical data depth using extreme value statistics

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q892256)