How deep is your model? Network topology selection from a model validation perspective

Simple item page
dc.contributor.authorNowaczyk, Nikolai
dc.contributor.authorKienitz, Jörg
dc.contributor.authorAcar, Sarp K
dc.contributor.authorLiang, Qian
dc.date.accessioned2022-04-06T06:35:08Z
dc.date.available2022-04-06T06:35:08Z
dc.date.issued2022-01-03
dc.date.updated2022-01-09T04:09:14Z
dc.description.abstractDeep learning is a powerful tool, which is becoming increasingly popular in financial modeling. However, model validation requirements such as SR 11-7 pose a significant obstacle to the deployment of neural networks in a bank’s production system. Their typically high number of (hyper-)parameters poses a particular challenge to model selection, benchmarking and documentation. We present a simple grid based method together with an open source implementation and show how this pragmatically satisfies model validation requirements. We illustrate the method by learning the option pricing formula in the Black–Scholes and the Heston model.en_US
dc.identifier.apacitationNowaczyk, N., Kienitz, J., Acar, S. K., & Liang, Q. (2022). How deep is your model? Network topology selection from a model validation perspective. <i>Journal of Mathematics in Industry</i>, 12(1), 1. http://hdl.handle.net/11427/36275en_ZA
dc.identifier.chicagocitationNowaczyk, Nikolai, Jörg Kienitz, Sarp K Acar, and Qian Liang "How deep is your model? Network topology selection from a model validation perspective." <i>Journal of Mathematics in Industry</i> 12, 1. (2022): 1. http://hdl.handle.net/11427/36275en_ZA
dc.identifier.citationNowaczyk, N., Kienitz, J., Acar, S.K. & Liang, Q. 2022. How deep is your model? Network topology selection from a model validation perspective. <i>Journal of Mathematics in Industry.</i> 12(1):1. http://hdl.handle.net/11427/36275en_ZA
dc.identifier.ris TY - Journal Article AU - Nowaczyk, Nikolai AU - Kienitz, Jörg AU - Acar, Sarp K AU - Liang, Qian AB - Deep learning is a powerful tool, which is becoming increasingly popular in financial modeling. However, model validation requirements such as SR 11-7 pose a significant obstacle to the deployment of neural networks in a bank’s production system. Their typically high number of (hyper-)parameters poses a particular challenge to model selection, benchmarking and documentation. We present a simple grid based method together with an open source implementation and show how this pragmatically satisfies model validation requirements. We illustrate the method by learning the option pricing formula in the Black–Scholes and the Heston model. DA - 2022-01-03 DB - OpenUCT DP - University of Cape Town IS - 1 J1 - Journal of Mathematics in Industry KW - Neural networks KW - Model validation KW - SR 11-7 KW - Derivatives KW - Risk management KW - Pricing LK - https://open.uct.ac.za PY - 2022 T1 - How deep is your model? Network topology selection from a model validation perspective TI - How deep is your model? Network topology selection from a model validation perspective UR - http://hdl.handle.net/11427/36275 ER - en_ZA
dc.identifier.urihttps://doi.org/10.1186/s13362-021-00116-5
dc.identifier.urihttp://hdl.handle.net/11427/36275
dc.identifier.vancouvercitationNowaczyk N, Kienitz J, Acar SK, Liang Q. How deep is your model? Network topology selection from a model validation perspective. Journal of Mathematics in Industry. 2022;12(1):1. http://hdl.handle.net/11427/36275.en_ZA
dc.language.isoenen_US
dc.language.rfc3066en
dc.publisher.departmentAfrican Inst. of Fin. Markets and Risk Mngnten_US
dc.publisher.facultyFaculty of Commerceen_US
dc.rights.holderThe Author(s)
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/en_US
dc.sourceJournal of Mathematics in Industryen_US
dc.source.journalissue1en_US
dc.source.journalvolume12en_US
dc.source.pagination1en_US
dc.source.urihttps://www.hindawi.com/journals/jim/
dc.subjectNeural networksen_US
dc.subjectModel validationen_US
dc.subjectSR 11-7en_US
dc.subjectDerivativesen_US
dc.subjectRisk managementen_US
dc.subjectPricingen_US
dc.titleHow deep is your model? Network topology selection from a model validation perspectiveen_US
dc.typeJournal Articleen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
13362_2021_Article_116.pdf
Size:
2.89 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
0 B
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Journal Articles