A preliminary investigation of overfitting in evolutionary driven model induction : implications for financial modelling

DC FieldValueLanguage
dc.contributor.authorTuite, Cliodhna
dc.contributor.authorAgapitos, Alexandros
dc.contributor.authorO'Neill, Michael
dc.contributor.authorBrabazon, Anthony
dc.date.accessioned2011-08-02T16:28:31Z
dc.date.available2011-08-02T16:28:31Z
dc.date.copyrightSpringer-Verlag Berlin Heidelberg 2011en
dc.date.issued2011-04
dc.identifier.isbn978-3-642-20519-4
dc.identifier.urihttp://hdl.handle.net/10197/3059
dc.descriptionEvoStar 2011, 27-29 April, 2011, Torino Italyen
dc.description.abstractThis paper investigates the effects of early stopping as a method to counteract overfitting in evolutionary data modelling using Genetic Programming. Early stopping has been proposed as a method to avoid model overtraining, which has been shown to lead to a significant degradation of out-of-sample performance. If we assume some sort of performance metric maximisation, the most widely used early training stopping criterion is the moment within the learning process that an unbiased estimate of the performance of the model begins to decrease after a strictly monotonic increase through the earlier learning iterations. We are conducting an initial investigation on the effects of early stopping in the performance of Genetic Programming in symbolic regression and financial modelling. Empirical results suggest that early stopping using the above criterion increases the extrapolation abilities of symbolic regression models, but is by no means the optimal training-stopping criterion in the case of a real-world financial dataset.en
dc.description.sponsorshipScience Foundation Irelanden
dc.format.extent654963 bytes
dc.format.mimetypeapplication/pdf
dc.language.isoenen
dc.publisherSpringeren
dc.relation.ispartofDi Chio, C. et al (eds.). Applications of Evolutionary Computation EvoApplications 2011: EvoCOMNET, EvoFIN, EvoHOT, EvoMUSART, EvoSTIM, and EvoTRANSLOG, Torino, Italy, April 27-29, 2011, Proceedings, Part IIen
dc.relation.requiresBusiness Research Collectionen
dc.relation.requiresCASL Research Collectionen
dc.relation.requiresFMC² Research Collectionen
dc.subjectOverfittingen
dc.subjectEvolutionary data modellingen
dc.subjectGenetic programmingen
dc.subject.lcshEvolutionary computationen
dc.subject.lcshGenetic programming (Computer science)en
dc.subject.lcshFinance--Computer simulationen
dc.titleA preliminary investigation of overfitting in evolutionary driven model induction : implications for financial modellingen
dc.typeConference Publicationen
dc.internal.availabilityFull text availableen
dc.internal.webversionshttp://dx.doi.org/10.1007/978-3-642-20520-0_13-
dc.statusPeer revieweden
dc.identifier.doi10.1007/978-3-642-20520-0_13-
dc.neeo.contributorTuite|Cliodhna|aut|-
dc.neeo.contributorAgapitos|Alexandros|aut|-
dc.neeo.contributorO'Neill|Michael|aut|-
dc.neeo.contributorBrabazon|Anthony|aut|-
dc.description.admin12M embargo: release in April 2012 - AV 2/08/2011en
item.grantfulltextopen-
item.fulltextWith Fulltext-
Appears in Collections:FMC² Research Collection
Business Research Collection
CASL Research Collection
Files in This Item:
File Description SizeFormat 
Preliminary Investigation of Overfitting.pdf639.61 kBAdobe PDFDownload
Show simple item record

SCOPUSTM   
Citations 20

11
Last Week
0
Last month
checked on Mar 20, 2019

Page view(s) 20

158
checked on May 25, 2018

Download(s) 50

186
checked on May 25, 2018

Google ScholarTM

Check

Altmetric


This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.