A Sentiment-Informed Algorithmic Cryptocurrency Trading Strategy Validated Across Diverse Market Conditions
IEEE Access · 2026 · 14 · 71309-71333
DOI: 10.1109/access.2026.3691381 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Rentaro Noguchi und 3 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Cryptocurrency markets are highly volatile, making short-term price prediction challenging. Because cryptocurrency prices are largely driven by supply and demand, prior work has explored social media sentiment to predict price changes, typically finding weak correlations. Reported successes usually come from richer frameworks such as those combining sentiment with financial indicators, making it unclear whether gains arise from sentiment signals alone. Prior studies are often limited to short windows under one year; although they demonstrate success in specific market conditions, such studies often fail to account for year-to-year changes in market behavior. We developed a rule-based sentiment-informed trading algorithm using signals extracted from Reddit comments on six cryptocurrencies from 2018 to 2025. Our algorithm, Cryptocurrency Algorithm using Relative Volume Sentiment (CARVS), achieved a return of 72,240.4% and a Sortino ratio of 3.035 over eight years, compared with 1,235.9% and 1.216 for a buy-and-hold baseline. CARVS also outperformed a raw sentiment polarity baseline, which replaced the Relative Volume Sentiment signal with unprocessed sentiment polarity scores to isolate the contribution of our signal construction and noise reduction methods. This baseline produced a return of -15.2% and a Sortino ratio of 0.748. All reported results are net of transaction costs. These results were validated over an out-of-sample period of 2024–2025, in which CARVS generated statistically significant risk-adjusted excess returns over both baselines. Evaluated over the longest available historical period for the selected cryptocurrencies, these results suggest that properly handled social media sentiment can serve as an independent, effective trading signal in cryptocurrency markets.
Zitieren
APA
Noguchi, R., Mahadevu, A., Aggarwal, A., & Qin, H. (2026). A Sentiment-Informed Algorithmic Cryptocurrency Trading Strategy Validated Across Diverse Market Conditions. IEEE Access, 14, 71309-71333. https://doi.org/10.1109/access.2026.3691381
Vancouver
Noguchi R, Mahadevu A, Aggarwal A, Qin H. A Sentiment-Informed Algorithmic Cryptocurrency Trading Strategy Validated Across Diverse Market Conditions. IEEE Access. 2026;14:71309-71333. doi:10.1109/access.2026.3691381
BibTeX
@article{Noguchi2026,
author = {Noguchi, Rentaro and Mahadevu, Ashrita and Aggarwal, Aariv and Qin, Huifang},
title = {A Sentiment-Informed Algorithmic Cryptocurrency Trading Strategy Validated Across Diverse Market Conditions},
journal = {IEEE Access},
year = {2026},
volume = {14},
pages = {71309--71333},
doi = {10.1109/access.2026.3691381}
}
EUVIMED