Man versus Machine Learning Revisited
The Review of Financial Studies · 01.12.2025 · 38 (12) · 3768-3790
DOI: 10.1093/rfs/hhaf066 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Review of Financial Studies, online erschienen am 02.09.2025. Verfasst von Yingguang Zhang und 2 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 30 Literaturangaben.
Abstract (Originalsprache)
Binsbergen, Han, and Lopez-Lira (2023) predict analysts’ forecast errors using a random forest model. A strategy that trades against this model’s predictions earns a monthly alpha of 1.54% ($ t $-value = 5.84). This estimate represents a large improvement over studies using classical statistical methods. We attribute the difference to a look-ahead bias. Removing the bias erases the alpha. Linear models yield as accurate forecasts and superior trading profits. Neither alternative machine learning models nor combinations thereof resurrect the predictability. We discuss the state of research into the term structure of analysts’ forecasts and its causal relationship with returns.
Zitieren
APA
Zhang, Y., Zhu, Y., & Linnainmaa, J. T. (2025). Man versus Machine Learning Revisited. The Review of Financial Studies, 38(12), 3768-3790. https://doi.org/10.1093/rfs/hhaf066
Vancouver
Zhang Y, Zhu Y, Linnainmaa JT. Man versus Machine Learning Revisited. The Review of Financial Studies. 2025;38(12):3768-3790. doi:10.1093/rfs/hhaf066
BibTeX
@article{Zhang2025,
author = {Zhang, Yingguang and Zhu, Yandi and Linnainmaa, Juhani T},
title = {Man versus Machine Learning Revisited},
journal = {The Review of Financial Studies},
year = {2025},
volume = {38},
number = {12},
pages = {3768--3790},
doi = {10.1093/rfs/hhaf066}
}
EUVIMED