When Relevance Is Not Enough: Enhancing Financial Models through Redundancy-Aware Feature Selection
The Journal of Financial Data Science · 30.04.2026 · 8 (2) · 53-72
DOI: 10.3905/jfds.2026.003 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Journal of Financial Data Science, online erschienen am 18.03.2026. Verfasst von Mohamed Chelly und 3 weiteren Autor:innen.
Abstract (Originalsprache)
This article examines the trade-off between feature relevance and redundancy in feature selection for linear regression models. The analysis focuses on two widely used feature selection paradigms, F-test and random forest, each implemented in two variants: a relevance-only version and a redundancy-aware extension that jointly accounts for both factors. To evaluate these methods, the authors construct a synthetic data generation framework that enables precise control over average feature relevance and redundancy. By assessing out-of-sample performance across a grid of such configurations, they identify the conditions under which redundancy-aware methods offer meaningful advantages. These findings are further validated on three real-world financial datasets. Results show that relying on relevance alone is often insufficient, and explicitly addressing redundancy improves both the predictive performance and the explanatory power of linear models.
Zitieren
APA
Chelly, M., Egger, L., Linger, M., & Boulanger, G. (2026). When Relevance Is Not Enough: Enhancing Financial Models through Redundancy-Aware Feature Selection. The Journal of Financial Data Science, 8(2), 53-72. https://doi.org/10.3905/jfds.2026.003
Vancouver
Chelly M, Egger L, Linger M, Boulanger G. When Relevance Is Not Enough: Enhancing Financial Models through Redundancy-Aware Feature Selection. The Journal of Financial Data Science. 2026;8(2):53-72. doi:10.3905/jfds.2026.003
BibTeX
@article{Chelly2026,
author = {Chelly, Mohamed and Egger, Lorena and Linger, Mathis and Boulanger, Guillaume},
title = {When Relevance Is Not Enough: Enhancing Financial Models through Redundancy-Aware Feature Selection},
journal = {The Journal of Financial Data Science},
year = {2026},
volume = {8},
number = {2},
pages = {53--72},
doi = {10.3905/jfds.2026.003}
}
EUVIMED