Robust Singular Value Decompositions: A New Approach to Projection Pursuit

DOI10.2307/2290330MaRDI QIDQ5288908zbMATH OpenOpenAlexFDO

Authors Larry P. Ammann

Publication date 28 October 1993

Full work available at URL https://doi.org/10.2307/2290330

zbMATH Keywords

exploratory data analysiseigenvaluesrobust covariance estimationsingular value decompositiondiscriminant analysisMonte Carlo studyrotationrobust estimatesmultivariate outliersexamplesnew algorithmrobust location estimatorsrobust principal componentslarge sample sizeserrors-in- variables regressioncovariance matrix eigenvectorsgeneral \(M\) estimationGM estimation problemhigh- dimensional dataleast squares regression fitregression hyperplanesrobust regression problems

Mathematics Subject Classification ID

Multivariate analysis (62H99) Probabilistic methods, stochastic differential equations (65C99) Estimation in multivariate analysis (62H12) Robustness and adaptive procedures (parametric inference) (62F35) Numerical computation of eigenvalues and eigenvectors of matrices (65F15)

Recommendations

• Robust eigenspace assignment using singular value sensitivities
• Reduced conservative singular value analysis for robustness
• A New Approach to Generalized Singular Value Decomposition
• Two proposals for robust PCA using semidefinite programming
• A new source of structured singular value decomposition problems
• Robust PCA and pairs of projections in a Hilbert space
• Robust eigenstructure assignment by a projection method - Applications using multiple optimization criteria

Cited in

(9)

• Robust computation of linear models by convex relaxation
• Statistical inference based on robust low-rank data matrix approximation
• Robust regularized singular value decomposition with application to mortality data
• Inferences based on multiple skipped correlations
• Robust singular value decomposition with application to video surveillance background modelling
• The percentage bend correlation coefficient
• On principal component analysis in \(L_{1}\).
• High breakdown estimators for principal components: the projection-pursuit approach revis\-ited
• Robust Padé Approximation via SVD

This page was built for publication: Robust Singular Value Decompositions: A New Approach to Projection Pursuit

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5288908)