Simple item page
 

On estimation of a partitioned covariance matrix with linearly structured blocks

cris.lastimport.scopus2025-10-23T06:55:23Z
cris.virtual.author-orcid0000-0001-5473-3419
cris.virtual.author-orcid0000-0001-9675-973X
cris.virtual.author-orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtual.author-orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtualsource.author-orcide9e0b41e-e4f0-4842-94d4-95738eaea207
cris.virtualsource.author-orcid1fe9b477-cfe7-4f84-8ce8-623637148ff1
cris.virtualsource.author-orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtualsource.author-orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
dc.abstract.enThe aim of this paper is to introduce an estimation method for a linearly structured partitioned covariance matrix. In contrast to well known linear structures of partitioned matrices, for example block compound symmetry, we allow the diagonal blocks of the covariance matrix to be of different dimensions. We adapt the shrinkage method to improve the properties of the projection of the sample covariance matrix onto the linear structure space. As spaces of target matrices, we choose various quadratic subspaces of structure space. This is a novel approach in the context of the structure space under consideration, and as a result a positive definite and well-conditioned estimator having the desired structure is determined. It is also shown that the statistical and algebraic properties of the estimator depend on the choice of target space.
dc.affiliationWydział Rolnictwa, Ogrodnictwa i Biotechnologii
dc.affiliation.instituteKatedra Metod Matematycznych i Statystycznych
dc.contributor.authorFilipiak, Katarzyna
dc.contributor.authorMarkiewicz, Augustyn
dc.contributor.authorMieldzioc, Adam
dc.contributor.authorMrowińska, Malwina
dc.date.access2025-06-10
dc.date.accessioned2025-07-03T07:03:29Z
dc.date.available2025-07-03T07:03:29Z
dc.date.copyright2025-05-14
dc.date.issued2025
dc.description.accesstimeat_publication
dc.description.bibliographyil., bibliogr.
dc.description.financepublication_nocost
dc.description.financecost0,00
dc.description.if1,1
dc.description.number4
dc.description.points100
dc.description.versionfinal_published
dc.description.volume66
dc.identifier.doi10.1007/s00362-025-01718-6
dc.identifier.eissn1613-9798
dc.identifier.issn0932-5026
dc.identifier.urihttps://sciencerep.up.poznan.pl/handle/item/3809
dc.identifier.weblinkhttps://link.springer.com/article/10.1007/s00362-025-01718-6
dc.languageen
dc.relation.ispartofStatistical Papers
dc.relation.pagesart. 98
dc.rightsCC-BY-NC-ND
dc.sciencecloudsend
dc.share.typeOTHER
dc.subject.enmultivariate model
dc.subject.encovariance structure
dc.subject.enleast squares estimation
dc.subject.enprojection
dc.subject.enshrinking
dc.subject.enquadratic subspace
dc.titleOn estimation of a partitioned covariance matrix with linearly structured blocks
dc.typeJournalArticle
dspace.entity.typePublication
oaire.citation.issue4
oaire.citation.volume66