High breakdown-point estimates of regression by means of the minimization of an efficient scale
A new class of robust estimates, τ estimates, is introduced. The estimates have simultaneously the following properties: (a) they are qualitatively robust, (b) their breakdown point is .5, and (c) they are highly efficient for regression models with normal errors. They are defined by minimizing a ne...
Guardado en:
Publicado: |
1988
|
---|---|
Materias: | |
Acceso en línea: | https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_01621459_v83_n402_p406_Yohai http://hdl.handle.net/20.500.12110/paper_01621459_v83_n402_p406_Yohai |
Aporte de: |
id |
paper:paper_01621459_v83_n402_p406_Yohai |
---|---|
record_format |
dspace |
spelling |
paper:paper_01621459_v83_n402_p406_Yohai2023-06-08T15:13:39Z High breakdown-point estimates of regression by means of the minimization of an efficient scale Bias robustness High efficiency Robust estimate A new class of robust estimates, τ estimates, is introduced. The estimates have simultaneously the following properties: (a) they are qualitatively robust, (b) their breakdown point is .5, and (c) they are highly efficient for regression models with normal errors. They are defined by minimizing a new scale estimate, τ, applied to the residuals. Asymptotically, a τ estimate is equivalent to an M estimate with a ψ function given by a weighted average of two ψ functions, one corresponding to a very robust estimate and the other to a highly efficient estimate. The weights are adaptive and depend on the underlying error distribution. We prove consistency and asymptotic normality and give a convergent iterative computing algorithm. Finally, we compare the biases produced by gross error contamination in the τ estimates and optimal bounded-influence estimates. © 1976 Taylor & Francis Group, LLC. 1988 https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_01621459_v83_n402_p406_Yohai http://hdl.handle.net/20.500.12110/paper_01621459_v83_n402_p406_Yohai |
institution |
Universidad de Buenos Aires |
institution_str |
I-28 |
repository_str |
R-134 |
collection |
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) |
topic |
Bias robustness High efficiency Robust estimate |
spellingShingle |
Bias robustness High efficiency Robust estimate High breakdown-point estimates of regression by means of the minimization of an efficient scale |
topic_facet |
Bias robustness High efficiency Robust estimate |
description |
A new class of robust estimates, τ estimates, is introduced. The estimates have simultaneously the following properties: (a) they are qualitatively robust, (b) their breakdown point is .5, and (c) they are highly efficient for regression models with normal errors. They are defined by minimizing a new scale estimate, τ, applied to the residuals. Asymptotically, a τ estimate is equivalent to an M estimate with a ψ function given by a weighted average of two ψ functions, one corresponding to a very robust estimate and the other to a highly efficient estimate. The weights are adaptive and depend on the underlying error distribution. We prove consistency and asymptotic normality and give a convergent iterative computing algorithm. Finally, we compare the biases produced by gross error contamination in the τ estimates and optimal bounded-influence estimates. © 1976 Taylor & Francis Group, LLC. |
title |
High breakdown-point estimates of regression by means of the minimization of an efficient scale |
title_short |
High breakdown-point estimates of regression by means of the minimization of an efficient scale |
title_full |
High breakdown-point estimates of regression by means of the minimization of an efficient scale |
title_fullStr |
High breakdown-point estimates of regression by means of the minimization of an efficient scale |
title_full_unstemmed |
High breakdown-point estimates of regression by means of the minimization of an efficient scale |
title_sort |
high breakdown-point estimates of regression by means of the minimization of an efficient scale |
publishDate |
1988 |
url |
https://bibliotecadigital.exactas.uba.ar/collection/paper/document/paper_01621459_v83_n402_p406_Yohai http://hdl.handle.net/20.500.12110/paper_01621459_v83_n402_p406_Yohai |
_version_ |
1768541603949969408 |