Loading…

Robust estimation for linear regression with asymmetric errors

The authors propose a new class of robust estimators for the parameters of a regression model in which the distribution of the error terms belongs to a class of exponential families including the log-gamma distribution. These estimates, which are a natural extension of the MM-estimates for ordinary...

Full description

Saved in:
Bibliographic Details
Published in:Canadian journal of statistics 2005-12, Vol.33 (4), p.511-528
Main Authors: Bianco, Ana M., Ben, Marta Garcia, Yohai, Víctor J.
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c4074-7e08289942b50642ea62224b08734154292f37f080d1912b5e8ab471ceaed1ef3
cites cdi_FETCH-LOGICAL-c4074-7e08289942b50642ea62224b08734154292f37f080d1912b5e8ab471ceaed1ef3
container_end_page 528
container_issue 4
container_start_page 511
container_title Canadian journal of statistics
container_volume 33
creator Bianco, Ana M.
Ben, Marta Garcia
Yohai, Víctor J.
description The authors propose a new class of robust estimators for the parameters of a regression model in which the distribution of the error terms belongs to a class of exponential families including the log-gamma distribution. These estimates, which are a natural extension of the MM-estimates for ordinary regression, may combine simultaneously high asymptotic efficiency and a high breakdown point. The authors prove the consistency and derive the asymptotic normal distribution of these estimates. A Monte Carlo study allows them to assess the efficiency and robustness of these estimates for finite samples. /// Les auteurs proposent une nouvelle classe d'estimateurs robustes pour les paramètres d'un modèle de régression dont la loi des termes d'erreur appartient à une classe de familles exponentielles incluant la distribution log-gamma. Ces estimateurs, qui généralisent de façon naturelle les MM-estimateurs de la régression ordinaire, peuvent avoir à la fois une bonne efficacité asymptotique et un point de rupture élevé. Les auteurs en démontrent la convergence et la normalité asymptotique. Une étude de Monte-Carlo leur permet d'évaluer l'efficacité et la robustesse des estimateurs dans des échantillons de taille finie.
doi_str_mv 10.1002/cjs.5550330404
format article
fullrecord <record><control><sourceid>jstor_proqu</sourceid><recordid>TN_cdi_proquest_miscellaneous_36516315</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><jstor_id>25046199</jstor_id><sourcerecordid>25046199</sourcerecordid><originalsourceid>FETCH-LOGICAL-c4074-7e08289942b50642ea62224b08734154292f37f080d1912b5e8ab471ceaed1ef3</originalsourceid><addsrcrecordid>eNqFkE1v00AQhlcVSA2Ba2-VLA7cHGb2w-u9VIKIpqCIqhTU3lZrZ9w6ONmya6vNv-9GRkFw4TTSvs8zmn0ZO0GYIQB_X6_jTCkFQoAEecQmqKHMjVS3L9gEBJpcaS6P2asY1wBCIfIJO_vmqyH2GcW-3bi-9dus8SHr2i25kAW6CxTj_vWx7e8zF3ebDfWhrTMKwYf4mr1sXBfpze85ZT_OP32fX-TLy8Xn-YdlXkvQMtcEJS-NkbxSUEhOruCcywpKLSQqyQ1vhG6ghBUaTBCVrpIaa3K0QmrElL0b9z4E_2tIx9pNG2vqOrclP0QrCoWFQJXAt_-Aaz-EbbrN8lSSQqGKBM1GqA4-xkCNfQjp92FnEey-S5u6tH-6TIIZhce2o91_aDv_cv2Xezq669j7cHC5AlmgMSnPx7yNPT0dchd-2kILrezN14Ut4aNa4pWwC_EMXRSO_w</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>200251356</pqid></control><display><type>article</type><title>Robust estimation for linear regression with asymmetric errors</title><source>International Bibliography of the Social Sciences (IBSS)</source><source>Wiley</source><source>JSTOR Archival Journals</source><creator>Bianco, Ana M. ; Ben, Marta Garcia ; Yohai, Víctor J.</creator><creatorcontrib>Bianco, Ana M. ; Ben, Marta Garcia ; Yohai, Víctor J.</creatorcontrib><description>The authors propose a new class of robust estimators for the parameters of a regression model in which the distribution of the error terms belongs to a class of exponential families including the log-gamma distribution. These estimates, which are a natural extension of the MM-estimates for ordinary regression, may combine simultaneously high asymptotic efficiency and a high breakdown point. The authors prove the consistency and derive the asymptotic normal distribution of these estimates. A Monte Carlo study allows them to assess the efficiency and robustness of these estimates for finite samples. /// Les auteurs proposent une nouvelle classe d'estimateurs robustes pour les paramètres d'un modèle de régression dont la loi des termes d'erreur appartient à une classe de familles exponentielles incluant la distribution log-gamma. Ces estimateurs, qui généralisent de façon naturelle les MM-estimateurs de la régression ordinaire, peuvent avoir à la fois une bonne efficacité asymptotique et un point de rupture élevé. Les auteurs en démontrent la convergence et la normalité asymptotique. Une étude de Monte-Carlo leur permet d'évaluer l'efficacité et la robustesse des estimateurs dans des échantillons de taille finie.</description><identifier>ISSN: 0319-5724</identifier><identifier>EISSN: 1708-945X</identifier><identifier>DOI: 10.1002/cjs.5550330404</identifier><language>eng</language><publisher>Hoboken: Wiley-Blackwell</publisher><subject>Consistent estimators ; Distribution ; Error ; Errors ; Estimation ; Estimators ; Generalized linear model ; Linear regression ; Log-gamma regression ; M-estimates ; Mathematical models ; Maximum likelihood estimation ; Maximum likelihood estimators ; Monte Carlo simulation ; Outliers ; Parameter estimation ; Point estimators ; Preliminary estimates ; Regression analysis ; robust estimates ; Statistical methods ; Studies</subject><ispartof>Canadian journal of statistics, 2005-12, Vol.33 (4), p.511-528</ispartof><rights>Copyright 2005 Statistical Society of Canada/Société statistique du Canada</rights><rights>Copyright © 2005 Statistical Society of Canada</rights><rights>Copyright Statistical Society of Canada Dec 2005</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c4074-7e08289942b50642ea62224b08734154292f37f080d1912b5e8ab471ceaed1ef3</citedby><cites>FETCH-LOGICAL-c4074-7e08289942b50642ea62224b08734154292f37f080d1912b5e8ab471ceaed1ef3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.jstor.org/stable/pdf/25046199$$EPDF$$P50$$Gjstor$$H</linktopdf><linktohtml>$$Uhttps://www.jstor.org/stable/25046199$$EHTML$$P50$$Gjstor$$H</linktohtml><link.rule.ids>314,780,784,27924,27925,33223,33224,58238,58471</link.rule.ids></links><search><creatorcontrib>Bianco, Ana M.</creatorcontrib><creatorcontrib>Ben, Marta Garcia</creatorcontrib><creatorcontrib>Yohai, Víctor J.</creatorcontrib><title>Robust estimation for linear regression with asymmetric errors</title><title>Canadian journal of statistics</title><addtitle>Can J Statistics</addtitle><description>The authors propose a new class of robust estimators for the parameters of a regression model in which the distribution of the error terms belongs to a class of exponential families including the log-gamma distribution. These estimates, which are a natural extension of the MM-estimates for ordinary regression, may combine simultaneously high asymptotic efficiency and a high breakdown point. The authors prove the consistency and derive the asymptotic normal distribution of these estimates. A Monte Carlo study allows them to assess the efficiency and robustness of these estimates for finite samples. /// Les auteurs proposent une nouvelle classe d'estimateurs robustes pour les paramètres d'un modèle de régression dont la loi des termes d'erreur appartient à une classe de familles exponentielles incluant la distribution log-gamma. Ces estimateurs, qui généralisent de façon naturelle les MM-estimateurs de la régression ordinaire, peuvent avoir à la fois une bonne efficacité asymptotique et un point de rupture élevé. Les auteurs en démontrent la convergence et la normalité asymptotique. Une étude de Monte-Carlo leur permet d'évaluer l'efficacité et la robustesse des estimateurs dans des échantillons de taille finie.</description><subject>Consistent estimators</subject><subject>Distribution</subject><subject>Error</subject><subject>Errors</subject><subject>Estimation</subject><subject>Estimators</subject><subject>Generalized linear model</subject><subject>Linear regression</subject><subject>Log-gamma regression</subject><subject>M-estimates</subject><subject>Mathematical models</subject><subject>Maximum likelihood estimation</subject><subject>Maximum likelihood estimators</subject><subject>Monte Carlo simulation</subject><subject>Outliers</subject><subject>Parameter estimation</subject><subject>Point estimators</subject><subject>Preliminary estimates</subject><subject>Regression analysis</subject><subject>robust estimates</subject><subject>Statistical methods</subject><subject>Studies</subject><issn>0319-5724</issn><issn>1708-945X</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2005</creationdate><recordtype>article</recordtype><sourceid>8BJ</sourceid><recordid>eNqFkE1v00AQhlcVSA2Ba2-VLA7cHGb2w-u9VIKIpqCIqhTU3lZrZ9w6ONmya6vNv-9GRkFw4TTSvs8zmn0ZO0GYIQB_X6_jTCkFQoAEecQmqKHMjVS3L9gEBJpcaS6P2asY1wBCIfIJO_vmqyH2GcW-3bi-9dus8SHr2i25kAW6CxTj_vWx7e8zF3ebDfWhrTMKwYf4mr1sXBfpze85ZT_OP32fX-TLy8Xn-YdlXkvQMtcEJS-NkbxSUEhOruCcywpKLSQqyQ1vhG6ghBUaTBCVrpIaa3K0QmrElL0b9z4E_2tIx9pNG2vqOrclP0QrCoWFQJXAt_-Aaz-EbbrN8lSSQqGKBM1GqA4-xkCNfQjp92FnEey-S5u6tH-6TIIZhce2o91_aDv_cv2Xezq669j7cHC5AlmgMSnPx7yNPT0dchd-2kILrezN14Ut4aNa4pWwC_EMXRSO_w</recordid><startdate>200512</startdate><enddate>200512</enddate><creator>Bianco, Ana M.</creator><creator>Ben, Marta Garcia</creator><creator>Yohai, Víctor J.</creator><general>Wiley-Blackwell</general><general>Statistical Society of Canada</general><general>Wiley‐Blackwell</general><general>Wiley Subscription Services, Inc</general><scope>BSCLL</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>8BJ</scope><scope>8FD</scope><scope>FQK</scope><scope>H8D</scope><scope>JBE</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>200512</creationdate><title>Robust estimation for linear regression with asymmetric errors</title><author>Bianco, Ana M. ; Ben, Marta Garcia ; Yohai, Víctor J.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c4074-7e08289942b50642ea62224b08734154292f37f080d1912b5e8ab471ceaed1ef3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2005</creationdate><topic>Consistent estimators</topic><topic>Distribution</topic><topic>Error</topic><topic>Errors</topic><topic>Estimation</topic><topic>Estimators</topic><topic>Generalized linear model</topic><topic>Linear regression</topic><topic>Log-gamma regression</topic><topic>M-estimates</topic><topic>Mathematical models</topic><topic>Maximum likelihood estimation</topic><topic>Maximum likelihood estimators</topic><topic>Monte Carlo simulation</topic><topic>Outliers</topic><topic>Parameter estimation</topic><topic>Point estimators</topic><topic>Preliminary estimates</topic><topic>Regression analysis</topic><topic>robust estimates</topic><topic>Statistical methods</topic><topic>Studies</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Bianco, Ana M.</creatorcontrib><creatorcontrib>Ben, Marta Garcia</creatorcontrib><creatorcontrib>Yohai, Víctor J.</creatorcontrib><collection>Istex</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>International Bibliography of the Social Sciences (IBSS)</collection><collection>Technology Research Database</collection><collection>International Bibliography of the Social Sciences</collection><collection>Aerospace Database</collection><collection>International Bibliography of the Social Sciences</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts – Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>Canadian journal of statistics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Bianco, Ana M.</au><au>Ben, Marta Garcia</au><au>Yohai, Víctor J.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Robust estimation for linear regression with asymmetric errors</atitle><jtitle>Canadian journal of statistics</jtitle><addtitle>Can J Statistics</addtitle><date>2005-12</date><risdate>2005</risdate><volume>33</volume><issue>4</issue><spage>511</spage><epage>528</epage><pages>511-528</pages><issn>0319-5724</issn><eissn>1708-945X</eissn><abstract>The authors propose a new class of robust estimators for the parameters of a regression model in which the distribution of the error terms belongs to a class of exponential families including the log-gamma distribution. These estimates, which are a natural extension of the MM-estimates for ordinary regression, may combine simultaneously high asymptotic efficiency and a high breakdown point. The authors prove the consistency and derive the asymptotic normal distribution of these estimates. A Monte Carlo study allows them to assess the efficiency and robustness of these estimates for finite samples. /// Les auteurs proposent une nouvelle classe d'estimateurs robustes pour les paramètres d'un modèle de régression dont la loi des termes d'erreur appartient à une classe de familles exponentielles incluant la distribution log-gamma. Ces estimateurs, qui généralisent de façon naturelle les MM-estimateurs de la régression ordinaire, peuvent avoir à la fois une bonne efficacité asymptotique et un point de rupture élevé. Les auteurs en démontrent la convergence et la normalité asymptotique. Une étude de Monte-Carlo leur permet d'évaluer l'efficacité et la robustesse des estimateurs dans des échantillons de taille finie.</abstract><cop>Hoboken</cop><pub>Wiley-Blackwell</pub><doi>10.1002/cjs.5550330404</doi><tpages>18</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0319-5724
ispartof Canadian journal of statistics, 2005-12, Vol.33 (4), p.511-528
issn 0319-5724
1708-945X
language eng
recordid cdi_proquest_miscellaneous_36516315
source International Bibliography of the Social Sciences (IBSS); Wiley; JSTOR Archival Journals
subjects Consistent estimators
Distribution
Error
Errors
Estimation
Estimators
Generalized linear model
Linear regression
Log-gamma regression
M-estimates
Mathematical models
Maximum likelihood estimation
Maximum likelihood estimators
Monte Carlo simulation
Outliers
Parameter estimation
Point estimators
Preliminary estimates
Regression analysis
robust estimates
Statistical methods
Studies
title Robust estimation for linear regression with asymmetric errors
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-01T19%3A57%3A35IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-jstor_proqu&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Robust%20estimation%20for%20linear%20regression%20with%20asymmetric%20errors&rft.jtitle=Canadian%20journal%20of%20statistics&rft.au=Bianco,%20Ana%20M.&rft.date=2005-12&rft.volume=33&rft.issue=4&rft.spage=511&rft.epage=528&rft.pages=511-528&rft.issn=0319-5724&rft.eissn=1708-945X&rft_id=info:doi/10.1002/cjs.5550330404&rft_dat=%3Cjstor_proqu%3E25046199%3C/jstor_proqu%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c4074-7e08289942b50642ea62224b08734154292f37f080d1912b5e8ab471ceaed1ef3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=200251356&rft_id=info:pmid/&rft_jstor_id=25046199&rfr_iscdi=true