6533b83afe1ef96bd12a7108

RESEARCH PRODUCT

The affine equivariant sign covariance matrix: asymptotic behavior and efficiencies

Hannu OjaEsa OllilaChristophe Croux

subject

Statistics and ProbabilityCovariance functionaffine equivarianceinfluence functionMultivariate normal distributionrobustnessComputer Science::Human-Computer InteractionEfficiencyestimatorsEstimation of covariance matricesScatter matrixStatisticsAffine equivarianceApplied mathematicsCMA-ESMultivariate signCovariance and correlation matricesRobustnessmultivariate medianMathematicsprincipal componentsInfluence functionNumerical AnalysisMultivariate medianCovariance matrixcovariance and correlation matricesdiscriminant-analysisCovarianceComputer Science::Otherdispersion matricesefficiencyLaw of total covariancemultivariate locationtestsStatistics Probability and Uncertaintyeigenvectors and eigenvaluesEigenvectors and eigenvaluesmultivariate sign

description

We consider the affine equivariant sign covariance matrix (SCM) introduced by Visuri et al. (J. Statist. Plann. Inference 91 (2000) 557). The population SCM is shown to be proportional to the inverse of the regular covariance matrix. The eigenvectors and standardized eigenvalues of the covariance, matrix can thus be derived from the SCM. We also construct an estimate of the covariance and correlation matrix based on the SCM. The influence functions and limiting distributions of the SCM and its eigenvectors and eigenvalues are found. Limiting efficiencies are given in multivariate normal and t-distribution cases. The estimates are highly efficient in the multivariate normal case and perform better than estimates based on the sample covariance matrix for heavy-tailed distributions. Simulations confirmed these findings for finite-sample efficiencies. (C) 2003 Elsevier Science (USA). All rights reserved. ispartof: Journal of multivariate analysis vol:87 issue:2 pages:328-355 status: published

yearjournalcountryeditionlanguage
2003-11-01Journal of Multivariate Analysis
10.1016/s0047-259x(03)00045-9http://dx.doi.org/10.1016/S0047-259X(03)00045-9