Predicting Bitcoin Price Movements Using Deep Learning: A Comparative Study of LSTM and Transformer Models with Multi-Source Data
Main Article Content
Abstract
Accurate prediction of Bitcoin price movements remains a challenging yet critical task in the realm of cryptocurrency trading and financial analytics. This paper investigates the application of advanced deep learning techniques, specifically Long Short-Term Memory (LSTM) networks and Transformer architectures, to forecast Bitcoin prices by leveraging not only historical price and trading volume data but also supplementary information in- cluding social media sentiment scores and macroeconomic indicators. We conduct a compar- ative study assessing the predictive performance and practical applicability of these mod- els, demonstrating the advantages and limitations of incorporating alternative data sources alongside traditional features. Experimental results on real-world datasets reveal insights into model accuracy, directional prediction ability, and robustness, informing the design of effective cryptocurrency forecasting strategies.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0). To view a copy of this license, visit https://creativecommons.org/licenses/by/4.0/
References
I. Nasirtafreshi, “Forecasting cryptocurrency prices using recurrent neural network and long short-term memory,” Data & Knowledge Engineering, vol. 139, p. 102009, 2022.
J. Chu, S. Chan, S. Nadarajah, and J. Osterrieder, “Garch modelling of cryptocurrencies,” Journal of Risk and Financial Management, vol. 10, no. 4, p. 17, 2017.
B. Aygun and E. K. Gunay, “Comparison of statistical and machine learning algorithms for forecasting daily bitcoin returns,” Avrupa Bilim ve Teknoloji Dergisi, no. 21, pp. 444–454, 2021.
S. Makridakis, E. Spiliotis, and V. Assimakopoulos, “Statistical and machine learning fore- casting methods: Concerns and ways forward,” PloS one, vol. 13, no. 3, p. e0194889, 2018.
G. Lai, W.-C. Chang, Y. Yang, and H. Liu, “Modeling long-and short-term temporal pat- terns with deep neural networks,” in The 41st international ACM SIGIR conference on research & development in information retrieval, pp. 95–104, 2018.
J. F. Torres, D. Hadjout, A. Sebaa, F. Mart ́ınez-A ́lvarez, and A. Troncoso, “Deep learning for time series forecasting: a survey,” Big data, vol. 9, no. 1, pp. 3–21, 2021.
H. Zhou, S. Zhang, J. Peng, S. Zhang, J. Li, H. Xiong, and W. Zhang, “Informer: Beyond efficient transformer for long sequence time-series forecasting,” in Proceedings of the AAAI conference on artificial intelligence, vol. 35, pp. 11106–11115, 2021.
X.-Y. Liu, Z. Xia, H. Yang, J. Gao, D. Zha, M. Zhu, C. D. Wang, Z. Wang, and J. Guo, “Dynamic datasets and market environments for financial reinforcement learning,” Machine Learning, vol. 113, no. 5, pp. 2795–2839, 2024.
A. M. Ozbayoglu, M. U. Gudelek, and O. B. Sezer, “Deep learning for financial applications : A survey,” Papers, 2020.
T. Dierckx, J. Davis, and W. Schoutens, “Using machine learning and alternative data to predict movements in market risk,” 2019.
J. V. Critien, A. Gatt, and J. Ellul, “Bitcoin price change and trend prediction through twitter sentiment and data volume,” Financial Innovation, vol. 8, no. 1, pp. 1–20, 2022.
G. Haritha and S. N.B, “Cryptocurrency price prediction using twitter sentiment analysis,” ArXiv, vol. abs/2303.09397, 2023.
R. Aroussi, “yfinance: Yahoo finance market data downloader.” https://pypi.org/ project/yfinance/, 2020.