General projection-pursuit estimators for the common principal components model: Influence functions and Monte Carlo study

The common principal components (CPC) model for several groups of multivariate observations assumes equal principal axes but possibly different variances along these axes among the groups. Under a CPCs model, generalized projection-pursuit estimators are defined by using score functions on the dispe...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: Boente, Graciela Lina
Publicado: 2006
Materias:
Asymptotic variances
Common principal components
Partial influence function
Projection-pursuit
Robust estimation
Acceso en línea:https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_0047259X_v97_n1_p124_Boente
http://hdl.handle.net/20.500.12110/paper_0047259X_v97_n1_p124_Boente
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
id paper:paper_0047259X_v97_n1_p124_Boente
record_format dspace
spelling paper:paper_0047259X_v97_n1_p124_Boente2023-06-08T15:05:37Z General projection-pursuit estimators for the common principal components model: Influence functions and Monte Carlo study Boente, Graciela Lina Asymptotic variances Common principal components Partial influence function Projection-pursuit Robust estimation The common principal components (CPC) model for several groups of multivariate observations assumes equal principal axes but possibly different variances along these axes among the groups. Under a CPCs model, generalized projection-pursuit estimators are defined by using score functions on the dispersion measure considered. Their partial influence functions are obtained and asymptotic variances are derived from them. When the score function is taken equal to the logarithm, it is shown that, under a proportionality model, the eigenvector estimators are optimal in the sense of minimizing the asymptotic variance of the eigenvectors, for a given scale measure. © 2004 Elsevier Inc. All rights reserved. Fil:Boente, G. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. 2006 https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_0047259X_v97_n1_p124_Boente http://hdl.handle.net/20.500.12110/paper_0047259X_v97_n1_p124_Boente
institution Universidad de Buenos Aires
institution_str I-28
repository_str R-134
collection Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic Asymptotic variances
Common principal components
Partial influence function
Projection-pursuit
Robust estimation
spellingShingle Asymptotic variances
Common principal components
Partial influence function
Projection-pursuit
Robust estimation
Boente, Graciela Lina
General projection-pursuit estimators for the common principal components model: Influence functions and Monte Carlo study
topic_facet Asymptotic variances
Common principal components
Partial influence function
Projection-pursuit
Robust estimation
description The common principal components (CPC) model for several groups of multivariate observations assumes equal principal axes but possibly different variances along these axes among the groups. Under a CPCs model, generalized projection-pursuit estimators are defined by using score functions on the dispersion measure considered. Their partial influence functions are obtained and asymptotic variances are derived from them. When the score function is taken equal to the logarithm, it is shown that, under a proportionality model, the eigenvector estimators are optimal in the sense of minimizing the asymptotic variance of the eigenvectors, for a given scale measure. © 2004 Elsevier Inc. All rights reserved.
author Boente, Graciela Lina
author_facet Boente, Graciela Lina
author_sort Boente, Graciela Lina
title General projection-pursuit estimators for the common principal components model: Influence functions and Monte Carlo study
title_short General projection-pursuit estimators for the common principal components model: Influence functions and Monte Carlo study
title_full General projection-pursuit estimators for the common principal components model: Influence functions and Monte Carlo study
title_fullStr General projection-pursuit estimators for the common principal components model: Influence functions and Monte Carlo study
title_full_unstemmed General projection-pursuit estimators for the common principal components model: Influence functions and Monte Carlo study
title_sort general projection-pursuit estimators for the common principal components model: influence functions and monte carlo study
publishDate 2006
url https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_0047259X_v97_n1_p124_Boente
http://hdl.handle.net/20.500.12110/paper_0047259X_v97_n1_p124_Boente
work_keys_str_mv AT boentegracielalina generalprojectionpursuitestimatorsforthecommonprincipalcomponentsmodelinfluencefunctionsandmontecarlostudy
_version_ 1768542451976372224

Ejemplares similares