Intelligent forecasting in bitcoin markets

Gil Cohen and Avishay Aiche

Finance Research Letters, 2025, vol. 71, issue C

Abstract: This paper examines the effectiveness of Artificial Intelligence (AI) in predicting Bitcoin's price movements. To achieve this, we developed two distinct trading strategies and compared their performance against each other and the traditional Buy and Hold (B&H) strategy. Over the period from January 2018 to September 2023, we found that the strategy optimized by ChatGPT 01-Preview, which integrates multiple technical indicators and sentiment analysis into a weighted composite index, delivered an exceptional total return of 944.85 %. The second strategy, that is using Extreme Gradient Boosting (XGBoost) technique achieved a total return of 189.05 %. The AI strategy's excess return of 755.8 % over the XGBoost strategy highlights the significant advantage of AI particularly in utilizing diverse data sources, such as social media, to predict Bitcoin's price trends more effectively than relying solely on economic data. Both trading strategies significantly outperformed the traditional B&H strategy, which returned 73.08 % over the same period. Furthermore, we found that AI has an advantage during periods of high Bitcoin price volatility.

Keywords: Bitcoin; AI; Machine learning; Random forest (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S1544612324015162
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:eee:finlet:v:71:y:2025:i:c:s1544612324015162

DOI: 10.1016/j.frl.2024.106487

Access Statistics for this article

Finance Research Letters is currently edited by R. Gençay

More articles in Finance Research Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().