Machine Learning Methods

Thomas Persson

Journal of Economics and Econometrics, 2025, vol. 68, issue 2, 106-129

Abstract: In high-dimensional regression settings with low signal-to-noise ratios, selecting an appropriate machine learning method is crucial for achieving reliable predictive performance. This study systematically evaluates several prevalent machine learning approaches, including regularised regression techniques (such as Lasso and Ridge), tree-based ensemble methods (such as Random Forest and Gradient Boosting), and neural networks. Through extensive simulations and real-world datasets, we assess their predictive accuracy, robustness to noise, and computational efficiency. Our findings provide insights into the relative strengths and weaknesses of these methods, offering practical guidelines for practitioners working with complex, high-dimensional data characterized by low signal-to-noise ratios.

Keywords: Machine learning; Lasso and Ridge; Random Forest; Gradient Boosting (search for similar items in EconPapers)
JEL-codes: C45 C52 C53 C55 C58 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://ideas.repec.org/a/eei/journl/v68y2025i2p106-129.html
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eei:journl:v:68:y:2025:i:2:p:106-129

Access Statistics for this article

More articles in Journal of Economics and Econometrics from Economics and Econometrics Society Contact information at EDIRC.
Bibliographic data for series maintained by Julia van Hove ().