Cross-validation Optimal Fold-Number for Model Selection
American Journal of Undergraduate Research · 28.12.2024 · 21 (3) · 15-29
DOI: 10.33697/ajur.2024.123 ↗Kurz gefasst
Fachartikel in der Zeitschrift American Journal of Undergraduate Research, online erschienen am 28.12.2024. Verfasst von Angelos Vasilopoulos und 1 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref).
Abstract (Originalsprache)
The resampling method of k-fold cross-validation is popular for error estimation and model selection in computational research. However, there is limited focus in the literature on the question of what fold number k is appropriate for various dataset dimensions. Here we review relevant literature and present a simulation of linear and least absolute shrinkage and selection operator (LASSO) regression prediction error estimation at various values of k and sample size n. In agreement with current literature, we find that contrary to a persisting understanding, there is no bias-variance trade-off in selection of k. Instead, with increasing k both bias and variance decrease, perhaps asymptotically. Our results also suggest a predictable relationship between optimal values of k and n. KEYWORDS: Cross-validation; Optimization; Fold number
Zitieren
APA
Vasilopoulos, A., & Matthews, G. (2024). Cross-validation Optimal Fold-Number for Model Selection. American Journal of Undergraduate Research, 21(3), 15-29. https://doi.org/10.33697/ajur.2024.123
Vancouver
Vasilopoulos A, Matthews G. Cross-validation Optimal Fold-Number for Model Selection. American Journal of Undergraduate Research. 2024;21(3):15-29. doi:10.33697/ajur.2024.123
BibTeX
@article{Vasilopoulos2024,
author = {Vasilopoulos, Angelos and Matthews, Gregory},
title = {Cross-validation Optimal Fold-Number for Model Selection},
journal = {American Journal of Undergraduate Research},
year = {2024},
volume = {21},
number = {3},
pages = {15--29},
doi = {10.33697/ajur.2024.123}
}
EUVIMED