EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

How Do Standard Time Series Forecasting Benchmarks Translate to Finance? A Broad Study of Risk Adjusted Performance

Adir Saly‐Kaufmann, Kieran Wood, Jan‐Peter Calliess, Stefan Zohren

Journal of Forecasting · 17.11.2026

DOI: 10.1002/for.70233 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Forecasting, online erschienen am 17.11.2026. Verfasst von Adir Saly‐Kaufmann und 3 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Forecasting financial asset returns is challenging due to low signal‐to‐noise ratios, nonstationarity and heavy‐tailed distributions. We benchmark 16 deep learning architectures for time series forecasting, including Transformers, State Space Models (SSMs), xLSTMs, and recurrent hybrids, on a 15‐year daily futures dataset using an end‐to‐end volatility‐targeted Sharpe ratio optimization task. Our results challenge prevailing assumptions in the literature. Several widely used “state‐of‐the‐art” architectures, including generic Transformers and SSMs, underperformed a simple AR(1) benchmark in a realistic financial forecasting task. The strongest results came from comparatively simple hybrid approaches using recurrent structures: a Variable Selection Network combined with an LSTM achieved the highest Sharpe ratio, while xLSTM showed the greatest robustness to transaction costs. These findings suggest that success on standard time‐series benchmarks does not reliably transfer to finance, and that architectures tailored to the unique properties of asset returns may remain more effective than generic attention‐based approaches for financial forecasting.

Zitieren

APA

Saly‐Kaufmann, A., Wood, K., Calliess, J., & Zohren, S. (2026). How Do Standard Time Series Forecasting Benchmarks Translate to Finance? A Broad Study of Risk Adjusted Performance. Journal of Forecasting. https://doi.org/10.1002/for.70233

Vancouver

Saly‐Kaufmann A, Wood K, Calliess J, Zohren S. How Do Standard Time Series Forecasting Benchmarks Translate to Finance? A Broad Study of Risk Adjusted Performance. Journal of Forecasting. 2026. doi:10.1002/for.70233

BibTeX
@article{SalyKaufmann2026,
  author = {Saly‐Kaufmann, Adir and Wood, Kieran and Calliess, Jan‐Peter and Zohren, Stefan},
  title = {How Do Standard Time Series Forecasting Benchmarks Translate to Finance? A Broad Study of Risk Adjusted Performance},
  journal = {Journal of Forecasting},
  year = {2026},
  doi = {10.1002/for.70233}
}
Kostenlos starten →