EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Beta Forecasting With Realized Beta Estimators and Machine Learning Algorithms

Bao Doan, Dulani Jayasuriya, John B. Lee, Jonathan J. Reeves

Journal of Forecasting · 2026-12 · 45 (8) · 3679-3709

DOI: 10.1002/for.70206 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Forecasting, online erschienen am 25.08.2026. Verfasst von Bao Doan und 3 weiteren Autor:innen. Die Arbeit nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper applies machine learning algorithms to the modeling of realized betas for the purposes of forecasting stock systematic risk. Higher levels of beta forecast accuracy are demonstrated, relative to other studies in the literature. These improvements are also highly significant, both statistically and economically. The approaches include individual machine learning algorithms and forecast combinations with multiple machine learning algorithms and realized betas. Forecast combinations lead to the highest level of beta forecast accuracy. The machine learning algorithms employed are ridge regression, decision tree learning, adaptive boosting, gradient boosting, random forests, and neural networks.

Zitieren

APA

Doan, B., Jayasuriya, D., Lee, J. B., & Reeves, J. J. (2026). Beta Forecasting With Realized Beta Estimators and Machine Learning Algorithms. Journal of Forecasting, 45(8), 3679-3709. https://doi.org/10.1002/for.70206

Vancouver

Doan B, Jayasuriya D, Lee JB, Reeves JJ. Beta Forecasting With Realized Beta Estimators and Machine Learning Algorithms. Journal of Forecasting. 2026;45(8):3679-3709. doi:10.1002/for.70206

BibTeX
@article{Doan2026,
  author = {Doan, Bao and Jayasuriya, Dulani and Lee, John B. and Reeves, Jonathan J.},
  title = {Beta Forecasting With Realized Beta Estimators and Machine Learning Algorithms},
  journal = {Journal of Forecasting},
  year = {2026},
  volume = {45},
  number = {8},
  pages = {3679--3709},
  doi = {10.1002/for.70206}
}
Kostenlos starten →