File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Modèles de réseaux de neurones pour l'analyse des séries temporelles ou la régression: Estimation, identification, méthode d'élagage SSM

TitleModèles de réseaux de neurones pour l'analyse des séries temporelles ou la régression: Estimation, identification, méthode d'élagage SSM
Authors
Rynkiewicz, JCottrell, MMangeas, MYao, JF
KeywordsAlmost sure identification
Asymptotic statistics
Statistical stepwise
Issue Date2001
PublisherLavoisier
Citation
Revue D'intelligence Artificielle, 2001, v. 15 n. 3-4, p. 317-332 How to Cite?
AbstractThis paper deals with neural network modeling for time series analysis or regression. Based on recent results about the least-square estimation for non-linear time series, we propose a complete and feasible methodology for both parameter estimation (learning process) and model selection (architecture selection). In particular, we solve the pruning problem for multilayer perceptron models with a stepwise search method by using a BIC criterion which is proved to be consistent.
Persistent Identifierhttp://hdl.handle.net/10722/132629
ISSN
0992-499X
2023 SCImago Journal Rankings: 0.278
References
References in Scopus

 

DC FieldValueLanguage
dc.contributor.authorRynkiewicz, Jen_HK
dc.contributor.authorCottrell, Men_HK
dc.contributor.authorMangeas, Men_HK
dc.contributor.authorYao, JFen_HK
dc.date.accessioned2011-03-28T09:27:06Z-
dc.date.available2011-03-28T09:27:06Z-
dc.date.issued2001en_HK
dc.identifier.citationRevue D'intelligence Artificielle, 2001, v. 15 n. 3-4, p. 317-332en_HK
dc.identifier.issn0992-499Xen_HK
dc.identifier.urihttp://hdl.handle.net/10722/132629-
dc.description.abstractThis paper deals with neural network modeling for time series analysis or regression. Based on recent results about the least-square estimation for non-linear time series, we propose a complete and feasible methodology for both parameter estimation (learning process) and model selection (architecture selection). In particular, we solve the pruning problem for multilayer perceptron models with a stepwise search method by using a BIC criterion which is proved to be consistent.en_HK
dc.languageengen_US
dc.publisherLavoisieren_US
dc.relation.ispartofRevue d'Intelligence Artificielleen_HK
dc.subjectAlmost sure identificationen_HK
dc.subjectAsymptotic statisticsen_HK
dc.subjectStatistical stepwiseen_HK
dc.titleModèles de réseaux de neurones pour l'analyse des séries temporelles ou la régression: Estimation, identification, méthode d'élagage SSMen_HK
dc.typeArticleen_HK
dc.identifier.emailYao, JF: jeffyao@hku.hken_HK
dc.identifier.authorityYao, JF=rp01473en_HK
dc.description.naturelink_to_subscribed_fulltexten_US
dc.identifier.scopuseid_2-s2.0-0042910500en_HK
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-0042910500&selection=ref&src=s&origin=recordpageen_HK
dc.identifier.volume15en_HK
dc.identifier.issue3-4en_HK
dc.identifier.spage317en_HK
dc.identifier.epage332en_HK
dc.publisher.placeFranceen_HK
dc.identifier.scopusauthoridRynkiewicz, J=8573566000en_HK
dc.identifier.scopusauthoridCottrell, M=7005442890en_HK
dc.identifier.scopusauthoridMangeas, M=6508084191en_HK
dc.identifier.scopusauthoridYao, JF=7403503451en_HK
dc.identifier.issnl0992-499X-

Export via OAI-PMH Interface in XML Formats

OR

Export to Other Non-XML Formats