Publikationstyp: Beitrag in wissenschaftlicher Zeitschrift
Art der Begutachtung: Peer review (Publikation)
Titel: Tackling the exponential scaling of signature-based generative adversarial networks for high-dimensional financial time-series generation
Autor/-in: De Meer Pardo, Fernando
Schwendner, Peter
Wunsch, Marcus
et. al: No
DOI: 10.3905/jfds.2022.1.109
Erschienen in: The Journal of Financial Data Science
Band(Heft): 4
Heft: 4
Seite(n): 110
Seiten bis: 132
Erscheinungsdatum: 2022
Verlag / Hrsg. Institution: Portfolio Management Research
ISSN: 2640-3943
Sprache: Englisch
Schlagwörter: GAN; Hierarchial clustering; Overfitting; Portfolio construction
Fachgebiet (DDC): 332: Finanzwirtschaft
Zusammenfassung: Generative 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.
Lizenz (gemäss Verlagsvertrag): Lizenz gemäss Verlagsvertrag
Departement: School of Engineering
School of Management and Law
Organisationseinheit: Institut für Datenanalyse und Prozessdesign (IDP)
Institut für Wealth & Asset Management (IWA)
Enthalten in den Sammlungen:Publikationen School of Engineering
Publikationen School of Management and Law

Dateien zu dieser Ressource:
Es gibt keine Dateien zu dieser Ressource.
Zur Langanzeige

Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt, soweit nicht anderweitig angezeigt.