Subspace rotations for high-dimensional outlier detection

DOI10.1016/j.jmva.2020.104713zbMath1464.62303OpenAlexW3116568734MaRDI QIDQ2022542

Jeongyoun Ahn, Hee Cheol Chung

Publication date: 29 April 2021

Published in: Journal of Multivariate Analysis (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.jmva.2020.104713

zbMATH Keywords

orthogonal groupStiefel manifoldrandomization testhigh dimensiongroup invariancelow sample size dataleft-spherical family

Mathematics Subject Classification ID

Hypothesis testing in multivariate analysis (62H15) Nonparametric statistical resampling methods (62G09)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Boundary behavior in high dimension, low sample size asymptotics of PCA
• Basic properties of strong mixing conditions. A survey and some open questions
• Outlier identification in high dimensions
• PCA consistency in high dimension, low sample size context
• Bayesian statistical inference on elliptical matrix distributions
• On singular Wishart and singular multivariate beta distributions
• Statistics on special manifolds
• Convergence of sample eigenvalues, eigenvectors, and principal component scores for ultra-high dimensional data
• Outlier Analysis
• On Strong Mixing Conditions for Stationary Gaussian Processes
• Outlier detection for high-dimensional data
• Rotation‐based multiple testing in the multivariate linear model
• Outlier detection for high dimensional data using the Comedian approach
• Distance-based outlier detection for high dimension, low sample size data
• The maximal data piling direction for discrimination
• Geometric Representation of High Dimension, Low Sample Size Data
• On the number of principal components in high dimensions
• The high-dimension, low-sample-size geometric representation holds under mild conditions
• Testing Statistical Hypotheses
• The Large-Sample Power of Tests Based on Permutations of Observations
• Normal Multivariate Analysis and the Orthogonal Group
• Clustering high dimension, low sample size data using the maximal data piling distance