Full metadata record
DC FieldValueLanguage
dc.contributor.authorDe Meer Pardo, Fernando-
dc.contributor.authorSchwendner, Peter-
dc.contributor.authorWunsch, Marcus-
dc.date.accessioned2021-11-29T08:57:41Z-
dc.date.available2021-11-29T08:57:41Z-
dc.date.issued2022-
dc.identifier.issn2640-3943de_CH
dc.identifier.issn2640-3951de_CH
dc.identifier.urihttps://digitalcollection.zhaw.ch/handle/11475/23562-
dc.description.abstractGenerative adversarial networks (GANs) have been shown to be able to generate samples of complex financial time series, particularly by employing the concept of path signatures, a universal description of the geometric properties of a data stream whose expected value uniquely characterizes the time series. Specifically, the SigCWGAN model (Ni et al. 2020) can generate time series of arbitrary length; however, the parameters of the neural network employed grow exponentially with the dimension of the underlying time series, which makes the model intractable when seeking to generate large financial market scenarios. To overcome this problem of dimensionality, the authors propose an iterative generation procedure relying on the concept of hierarchies in financial markets. The authors construct an ensemble of GANs that they call the Hierarchical-SigCWGAN, which is based on hierarchical clustering that approximates signatures in the spirit of the original model. The Hierarchical-SigCWGAN can scale to higher dimensions and generate large-dimensional scenarios in which the joint behavior of all the assets in the market is replicated. The model is validated by comparing its performance on a series of similarity metrics with respect to the original SigCWGAN on a dataset in which it is still tractable and by showing its scalability on a larger dataset.de_CH
dc.language.isoende_CH
dc.publisherPortfolio Management Researchde_CH
dc.relation.ispartofThe Journal of Financial Data Sciencede_CH
dc.rightsLicence according to publishing contractde_CH
dc.subjectGANde_CH
dc.subjectHierarchial clusteringde_CH
dc.subjectOverfittingde_CH
dc.subjectPortfolio constructionde_CH
dc.subject.ddc332: Finanzwirtschaftde_CH
dc.titleTackling the exponential scaling of signature-based generative adversarial networks for high-dimensional financial time-series generationde_CH
dc.typeBeitrag in wissenschaftlicher Zeitschriftde_CH
dcterms.typeTextde_CH
zhaw.departementSchool of Engineeringde_CH
zhaw.departementSchool of Management and Lawde_CH
zhaw.organisationalunitInstitut für Datenanalyse und Prozessdesign (IDP)de_CH
zhaw.organisationalunitInstitut für Wealth & Asset Management (IWA)de_CH
dc.identifier.doi10.3905/jfds.2022.1.109de_CH
zhaw.funding.euNode_CH
zhaw.issue4de_CH
zhaw.originated.zhawYesde_CH
zhaw.pages.end132de_CH
zhaw.pages.start110de_CH
zhaw.volume4de_CH
zhaw.publication.reviewPeer review (Publikation)de_CH
zhaw.author.additionalNode_CH
zhaw.display.portraitYesde_CH
Appears in collections:Publikationen School of Engineering
Publikationen School of Management and Law

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.