A fast algorithm for S-regression estimates

Equivariant high-breakdown point regression estimates are computationally expensive, and the corresponding algorithms become unfeasible for moderately large number of regressors. One important advance to improve the computational speed of one such estimator is the fast-LTS algorithm. This article pr...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Salibian-Barrera, M., Yohai, V.J.
Formato:	JOUR
Materias:	High breakdown point Linear regression Robustness
Acceso en línea:	http://hdl.handle.net/20.500.12110/paper_10618600_v15_n2_p414_SalibianBarrera
Aporte de:	Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires

id	todo:paper_10618600_v15_n2_p414_SalibianBarrera
record_format	dspace
spelling	todo:paper_10618600_v15_n2_p414_SalibianBarrera2023-10-03T16:01:09Z A fast algorithm for S-regression estimates Salibian-Barrera, M. Yohai, V.J. High breakdown point Linear regression Robustness Equivariant high-breakdown point regression estimates are computationally expensive, and the corresponding algorithms become unfeasible for moderately large number of regressors. One important advance to improve the computational speed of one such estimator is the fast-LTS algorithm. This article proposes an analogous algorithm for computing S-estimates. The new algorithm, that we call "fast-S", is also based on a "local improvement" step of the resampling initial candidates. This allows for a substantial reduction of the number of candidates required to obtain a good approximation to the optimal solution. We performed a simulation study which shows that S-estimators computed with the fast-S algorithm compare favorably to the LTS-estimators computed with the fast-LTS algorithm. ©2006 American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America. JOUR info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by/2.5/ar http://hdl.handle.net/20.500.12110/paper_10618600_v15_n2_p414_SalibianBarrera
institution	Universidad de Buenos Aires
institution_str	I-28
repository_str	R-134
collection	Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic	High breakdown point Linear regression Robustness
spellingShingle	High breakdown point Linear regression Robustness Salibian-Barrera, M. Yohai, V.J. A fast algorithm for S-regression estimates
topic_facet	High breakdown point Linear regression Robustness
description	Equivariant high-breakdown point regression estimates are computationally expensive, and the corresponding algorithms become unfeasible for moderately large number of regressors. One important advance to improve the computational speed of one such estimator is the fast-LTS algorithm. This article proposes an analogous algorithm for computing S-estimates. The new algorithm, that we call "fast-S", is also based on a "local improvement" step of the resampling initial candidates. This allows for a substantial reduction of the number of candidates required to obtain a good approximation to the optimal solution. We performed a simulation study which shows that S-estimators computed with the fast-S algorithm compare favorably to the LTS-estimators computed with the fast-LTS algorithm. ©2006 American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America.
format	JOUR
author	Salibian-Barrera, M. Yohai, V.J.
author_facet	Salibian-Barrera, M. Yohai, V.J.
author_sort	Salibian-Barrera, M.
title	A fast algorithm for S-regression estimates
title_short	A fast algorithm for S-regression estimates
title_full	A fast algorithm for S-regression estimates
title_fullStr	A fast algorithm for S-regression estimates
title_full_unstemmed	A fast algorithm for S-regression estimates
title_sort	fast algorithm for s-regression estimates
url	http://hdl.handle.net/20.500.12110/paper_10618600_v15_n2_p414_SalibianBarrera
work_keys_str_mv	AT salibianbarreram afastalgorithmforsregressionestimates AT yohaivj afastalgorithmforsregressionestimates AT salibianbarreram fastalgorithmforsregressionestimates AT yohaivj fastalgorithmforsregressionestimates
_version_	1782030097280139264

A fast algorithm for S-regression estimates

Ejemplares similares