File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: On the solution of the errors in variables problem using the l 1 norm

TitleOn the solution of the errors in variables problem using the l 1 norm
Authors
KeywordsAms Subject Classification: 65D10, 65K05
Issue Date1991
PublisherSpringer Verlag Dordrecht. The Journal's web site is located at http://springerlink.metapress.com/openurl.asp?genre=journal&issn=0006-3835
Citation
Bit, 1991, v. 31 n. 4, p. 697-710 How to Cite?
AbstractA fundamental problem in data analysis is that of fitting a given model to observed data. It is commonly assumed that only the dependent variable values are in error, and the least squares criterion is often used to fit the model. When significant errors occur in all the variables, then an alternative approach which is frequently suggested for this errors in variables problem is to minimize the sum of squared orthogonal distances between each data point and the curve described by the model equation. It has long been recognized that the use of least squares is not always satisfactory, and the l 1 criterion is often superior when estimating the true form of data which contain some very inaccurate observations. In this paper the measure of goodness of fit is taken to be the l 1 norm of the errors. A Levenberg-Marquardt method is proposed, and the main objective is to take full advantage of the structure of the subproblems so that they can be solved efficiently. © 1991 BIT Foundations.
Persistent Identifierhttp://hdl.handle.net/10722/155779
ISSN
2015 Impact Factor: 1.167
2015 SCImago Journal Rankings: 1.221
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorWatson, GAen_US
dc.contributor.authorYiu, KFCen_US
dc.date.accessioned2012-08-08T08:37:43Z-
dc.date.available2012-08-08T08:37:43Z-
dc.date.issued1991en_US
dc.identifier.citationBit, 1991, v. 31 n. 4, p. 697-710en_US
dc.identifier.issn0006-3835en_US
dc.identifier.urihttp://hdl.handle.net/10722/155779-
dc.description.abstractA fundamental problem in data analysis is that of fitting a given model to observed data. It is commonly assumed that only the dependent variable values are in error, and the least squares criterion is often used to fit the model. When significant errors occur in all the variables, then an alternative approach which is frequently suggested for this errors in variables problem is to minimize the sum of squared orthogonal distances between each data point and the curve described by the model equation. It has long been recognized that the use of least squares is not always satisfactory, and the l 1 criterion is often superior when estimating the true form of data which contain some very inaccurate observations. In this paper the measure of goodness of fit is taken to be the l 1 norm of the errors. A Levenberg-Marquardt method is proposed, and the main objective is to take full advantage of the structure of the subproblems so that they can be solved efficiently. © 1991 BIT Foundations.en_US
dc.languageengen_US
dc.publisherSpringer Verlag Dordrecht. The Journal's web site is located at http://springerlink.metapress.com/openurl.asp?genre=journal&issn=0006-3835en_US
dc.relation.ispartofBITen_US
dc.subjectAms Subject Classification: 65D10, 65K05en_US
dc.titleOn the solution of the errors in variables problem using the l 1 normen_US
dc.typeArticleen_US
dc.identifier.emailYiu, KFC:cedric@hkucc.hku.hken_US
dc.identifier.authorityYiu, KFC=rp00206en_US
dc.description.naturelink_to_subscribed_fulltexten_US
dc.identifier.doi10.1007/BF01933182en_US
dc.identifier.scopuseid_2-s2.0-0009257352en_US
dc.identifier.volume31en_US
dc.identifier.issue4en_US
dc.identifier.spage697en_US
dc.identifier.epage710en_US
dc.identifier.isiWOS:A1991HA99700012-
dc.publisher.placeNetherlandsen_US
dc.identifier.scopusauthoridWatson, GA=7401433832en_US
dc.identifier.scopusauthoridYiu, KFC=24802813000en_US

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats