High finite-sample efficiency and robustness based on distance-constrained maximum likelihood

Good robust estimators can be tuned to combine a high breakdown point and a specified asymptotic efficiency at a central model. This happens in regression with MM- and -estimators among others. However, the finite-sample efficiency of these estimators can be much lower than the asymptotic one. To ov...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Maronna, Ricardo Antonio, Yohai, Victor Jaime
Formato: Articulo Preprint
Lenguaje:Inglés
Publicado: 2015
Materias:
Matemática
Linear model
Robust estimator
High efficiency
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/101650
https://ri.conicet.gov.ar/11336/42723
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-101650
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Matemática
Linear model
Robust estimator
High efficiency
spellingShingle Matemática
Linear model
Robust estimator
High efficiency
Maronna, Ricardo Antonio
Yohai, Victor Jaime
High finite-sample efficiency and robustness based on distance-constrained maximum likelihood
topic_facet Matemática
Linear model
Robust estimator
High efficiency
description Good robust estimators can be tuned to combine a high breakdown point and a specified asymptotic efficiency at a central model. This happens in regression with MM- and -estimators among others. However, the finite-sample efficiency of these estimators can be much lower than the asymptotic one. To overcome this drawback, an approach is proposed for parametric models, which is based on a distance between parameters. Given a robust estimator, the proposed one is obtained by maximizing the likelihood under the constraint that the distance is less than a given threshold. For the linear model with normal errors, simulations show that the proposed estimator attains a finite-sample efficiency close to one while improving the robustness of the initial estimator. The same approach also shows good results in the estimation of multivariate location and scatter.
format Articulo
Preprint
author Maronna, Ricardo Antonio
Yohai, Victor Jaime
author_facet Maronna, Ricardo Antonio
Yohai, Victor Jaime
author_sort Maronna, Ricardo Antonio
title High finite-sample efficiency and robustness based on distance-constrained maximum likelihood
title_short High finite-sample efficiency and robustness based on distance-constrained maximum likelihood
title_full High finite-sample efficiency and robustness based on distance-constrained maximum likelihood
title_fullStr High finite-sample efficiency and robustness based on distance-constrained maximum likelihood
title_full_unstemmed High finite-sample efficiency and robustness based on distance-constrained maximum likelihood
title_sort high finite-sample efficiency and robustness based on distance-constrained maximum likelihood
publishDate 2015
url http://sedici.unlp.edu.ar/handle/10915/101650
https://ri.conicet.gov.ar/11336/42723
work_keys_str_mv AT maronnaricardoantonio highfinitesampleefficiencyandrobustnessbasedondistanceconstrainedmaximumlikelihood
AT yohaivictorjaime highfinitesampleefficiencyandrobustnessbasedondistanceconstrainedmaximumlikelihood
bdutipo_str Repositorios
_version_ 1764820442839777280

Ejemplares similares