Robust estimation for ARMA models

This paper introduces a new class of robust estimates for ARMA models. They are M-estimates, but the residuals are computed so the effect of one outlier is limited to the period where it occurs. These estimates are closely related to those based on a robust filter, but they have two important advant...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Muler, N., Peña, D., Yohai, V.J.
Formato: JOUR
Materias:
MM-estimates
Outliers
Time series
Acceso en línea:http://hdl.handle.net/20.500.12110/paper_00905364_v37_n2_p816_Muler
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
id todo:paper_00905364_v37_n2_p816_Muler
record_format dspace
spelling todo:paper_00905364_v37_n2_p816_Muler2023-10-03T14:54:42Z Robust estimation for ARMA models Muler, N. Peña, D. Yohai, V.J. MM-estimates Outliers Time series This paper introduces a new class of robust estimates for ARMA models. They are M-estimates, but the residuals are computed so the effect of one outlier is limited to the period where it occurs. These estimates are closely related to those based on a robust filter, but they have two important advantages: they are consistent and the asymptotic theory is tractable. We perform a Monte Carlo where we show that these estimates compare favorably with respect to standard M-estimates and to estimates based on a diagnostic procedure. © Institute of Mathematical Statistics, 2009. Fil:Muler, N. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. JOUR info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by/2.5/ar http://hdl.handle.net/20.500.12110/paper_00905364_v37_n2_p816_Muler
institution Universidad de Buenos Aires
institution_str I-28
repository_str R-134
collection Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic MM-estimates
Outliers
Time series
spellingShingle MM-estimates
Outliers
Time series
Muler, N.
Peña, D.
Yohai, V.J.
Robust estimation for ARMA models
topic_facet MM-estimates
Outliers
Time series
description This paper introduces a new class of robust estimates for ARMA models. They are M-estimates, but the residuals are computed so the effect of one outlier is limited to the period where it occurs. These estimates are closely related to those based on a robust filter, but they have two important advantages: they are consistent and the asymptotic theory is tractable. We perform a Monte Carlo where we show that these estimates compare favorably with respect to standard M-estimates and to estimates based on a diagnostic procedure. © Institute of Mathematical Statistics, 2009.
format JOUR
author Muler, N.
Peña, D.
Yohai, V.J.
author_facet Muler, N.
Peña, D.
Yohai, V.J.
author_sort Muler, N.
title Robust estimation for ARMA models
title_short Robust estimation for ARMA models
title_full Robust estimation for ARMA models
title_fullStr Robust estimation for ARMA models
title_full_unstemmed Robust estimation for ARMA models
title_sort robust estimation for arma models
url http://hdl.handle.net/20.500.12110/paper_00905364_v37_n2_p816_Muler
work_keys_str_mv AT mulern robustestimationforarmamodels
AT penad robustestimationforarmamodels
AT yohaivj robustestimationforarmamodels
_version_ 1807319510011609088

Ejemplares similares