A note on sensitivity of principal component subspaces and the efficient detection of influential observations in high dimensions

DOI10.1214/08-EJS201zbMath1320.62140arXiv0803.0402OpenAlexW2000327239MaRDI QIDQ1951757

Luke A. Prendergast

Publication date: 24 May 2013

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/0803.0402

zbMATH Keywords

perturbationprincipal component analysisinfluential observationsdistance between subspaces

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Estimation in multivariate analysis (62H12) Robustness and adaptive procedures (parametric inference) (62F35)

Related Items (6)

A new and practical influence measure for subsets of covariance matrix sample principal components with applications to high dimensional datasets ⋮ A concentration approach to sensitivity studies in statistical estimation problems ⋮ A correction of approximations used in sensitivity study of principal component analysis ⋮ Sensitivity of principal component subspaces: a comment on Prendergast's paper ⋮ Influence Functions for Dimension Reduction Methods: An Example Influence Study of Principal Hessian Direction Analysis ⋮ Comparative Study of Robust Estimators Based on a Sensitivity Coefficient in Principal Component Analysis

Uses Software

• LAPACK

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• On the relation between S-estimators and M-estimators of multivariate location and covariance
• Asymptotic behaviour of S-estimates of multivariate location parameters and dispersion matrices
• Asymptotics of reweighted estimators of multivariate location and scatter
• Regression analysis under link violation
• Influence function and efficiency of the minimum covariance determinant scatter matrix estimator
• Sensitivity of principal Hessian direction analysis
• DETECTING INFLUENTIAL OBSERVATIONS IN SLICED INVERSE REGRESSION ANALYSIS
• Influence in principal components analysis
• Principal Hessian Directions Revisited
• Quadratic perturbation expansions of certain functions of eigenvalues and eigenvectors and their application to sensitivity analysis in multivariate methods
• Sliced Inverse Regression for Dimension Reduction
• On Principal Hessian Directions for Data Visualization and Dimension Reduction: Another Application of Stein's Lemma
• The Influence Curve and Its Role in Robust Estimation
• Robust estimation and outlier detection with correlation coefficients
• The identification of a particular nonlinear time series system
• Principal component analysis based on robust estimators of the covariance or correlation matrix: influence functions and efficiencies
• Implications of influence function analysis for sliced inverse regression and sliced average variance estimation
• Influence Analysis in Principal Component Analysis Through Power-Series Expansions
• Comment

This page was built for publication: A note on sensitivity of principal component subspaces and the efficient detection of influential observations in high dimensions