B-matrices and its Applications to Linear Models

TitleB-matrices and its Applications to Linear Models
Publication TypeConference Paper
Year of Publication2015
AuthorsCarvalho F, Covas R
EditorSimos TE, Tsitouras C}
Conference NamePROCEEDINGS OF THE INTERNATIONAL CONFERENCE OF NUMERICAL ANALYSIS AND APPLIED MATHEMATICS 2014 (ICNAAM-2014)
PublisherAMER INST PHYSICS
Conference Location2 HUNTINGTON QUADRANGLE, STE 1NO1, MELVILLE, NY 11747-4501 USA
ISBN Number978-0-7354-1287-3
KeywordsAnalysis of variance, Inference, Jordan algebras, Linear models
Abstract

{A new class of matrices is introduced, with interesting properties. Their use can be of most importance and the discussion of a possible use in statistical inference is presented. Namely the use in a class of models, the ones with mean driven balance, which are also defined. We point out the relevance of such models, when dealing with mixed models and the approach, for inference proposes, is the use of commutative Jordan algebras of symmetric matrices. The properties of symmetric B-matrices take a decisive roll in this approach.}

DOI10.1063/1.4912417