Robustifying principal component analysis with spatial sign vectors

DSpace/Manakin Repository

Show simple item record Taskinen, Sara Koch, Inge Oja, Hannu 2012-11-30T12:47:16Z 2012-11-30T12:47:16Z 2012 fi
dc.identifier.citation Taskinen, S., Koch, I., & Oja, H. (2012). Robustifying principal component analysis with spatial sign vectors. <em>Statistics and Probability Letters</em>, 82 (4), 765-774. <a href="">doi:10.1016/j.spl.2012.01.001</a> fi
dc.identifier.issn 0167-7152
dc.identifier.other TUTKAID_50580
dc.description.abstract In this paper, we apply orthogonally equivariant spatial sign covariance matrices as well as their affine equivariant counterparts in principal component analysis. The influence functions and asymptotic covariance matrices of eigenvectors based on robust covariance estimators are derived in order to compare the robustness and efficiency properties. We show in particular that the estimators that use pairwise differences of the observed data have very good efficiency properties, providing practical robust alternatives to classical sample covariance matrix based methods. fi
dc.language.iso eng
dc.publisher Elsevier
dc.relation.ispartofseries Statistics and Probability Letters
dc.rights © Elsevier. This is a pre-print version of an article whose final and definitive form has been published by Elsevier.
dc.rights openAccess fi
dc.subject.other robustness fi
dc.subject.other Affiinisti ekvivarianttisuus
dc.subject.other tehokkuus
dc.subject.other vaikutusfunktio
dc.subject.other robustisuus
dc.subject.other spatiaalinen merkkivektori
dc.subject.other Affine equivariance
dc.subject.other efficiency
dc.subject.other influence function
dc.subject.other spatial sign vector
dc.title Robustifying principal component analysis with spatial sign vectors
dc.type Article en
dc.identifier.urn URN:NBN:fi:jyu-201211293124
dc.subject.kota 112
dc.contributor.laitos Matematiikan ja tilastotieteen laitos fi
dc.contributor.laitos Department of Mathematics and Statistics en
dc.contributor.oppiaine tilastotiede fi
dc.identifier.volume 82
dc.identifier.issue 4
jyx.tutka.pagetopage 765774
dc.identifier.doi 10.1016/j.spl.2012.01.001 2012-11-29T10:40:00Z
dc.description.version Pre-print version

This item appears in the following Collection(s)

Show simple item record