Multiple yield curve modeling and forecasting using deep learning

This manuscript introduces deep learning models that simultaneously describe the dynamics of several yield curves. We aim to learn the dependence structure among the different yield curves induced by the globalization of financial markets and exploit it to produce more accurate forecasts. By combini...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	ASTIN Bulletin : The Journal of the IAA Ročník 54; číslo 3; s. 463 - 494
Hlavní autoři:	Richman, Ronald, Scognamiglio, Salvatore
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York, USA Cambridge University Press 01.09.2024
Témata:	Asset liability management Deep learning Discount rates Economic forecasting Effectiveness Expected values Interest rates Liability Life insurance Machine learning Securities markets Solvency Valuation Yield curve Deep learning transfer learning asset–liability management interest rate risk Real-world modeling Solvency II, IFRS 17 multiple yield curve modeling attention models Nelson–Siegel model value-at-risk
ISSN:	0515-0361, 1783-1350
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	This manuscript introduces deep learning models that simultaneously describe the dynamics of several yield curves. We aim to learn the dependence structure among the different yield curves induced by the globalization of financial markets and exploit it to produce more accurate forecasts. By combining the self-attention mechanism and nonparametric quantile regression, our model generates both point and interval forecasts of future yields. The architecture is designed to avoid quantile crossing issues affecting multiple quantile regression models. Numerical experiments conducted on two different datasets confirm the effectiveness of our approach. Finally, we explore potential extensions and enhancements by incorporating deep ensemble methods and transfer learning mechanisms.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ISSN:	0515-0361 1783-1350
DOI:	10.1017/asb.2024.26