Newswire Tone-Overlay Commodity Portfolios
Fernandez-Perez, A., Fuertes, A-M. ORCID: 0000-0001-6468-9845, Miffre, J. & Zhao, N. (2025).
Newswire Tone-Overlay Commodity Portfolios.
Journal of Banking & Finance, 178,
article number 107501.
doi: 10.1016/j.jbankfin.2025.107501
Abstract
This paper introduces the tone-overlay framework for adjusting traditional commodity signals based on the level of salient optimism or pessimism in commodity newswires. By implementing the novel tone-overlay allocation strategy on 26 commodities using traditional allocation signals, we demonstrate that the resulting long-short portfolios yield substantial performance gains compared to the corresponding plain-vanilla traditional portfolios. Our findings suggest that newswire tone provides short-term predictive power for commodity futures returns, beyond well-known commodity characteristics. The tone-overlay portfolios harness a temporary mispricing that reflects an overreaction of commodity futures prices to commodity-specific newswire tone. The outperformance of the tone overlay strengthens with the salience of the newswire tone, consistent with theories of limited investor attention.
Publication Type: | Article |
---|---|
Additional Information: | © 2025. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/ |
Publisher Keywords: | Newswire tone-overlay portfolio, Textual analysis, Sentiment, Commodity futures, Tactical allocation, Mispricing, Salience |
Subjects: | H Social Sciences > HE Transportation and Communications H Social Sciences > HG Finance |
Departments: | Bayes Business School Bayes Business School > Finance |
SWORD Depositor: |
![[thumbnail of SSRN_Newswire tone overlay comm port JBF2025.pdf]](https://openaccess.city.ac.uk/style/images/fileicons/text.png)
This document is not freely accessible until 21 December 2026 due to copyright restrictions.
Available under License Creative Commons Attribution Non-commercial No Derivatives.
To request a copy, please use the button below.
Request a copyExport
Downloads
Downloads per month over past year