Forecasting Crypto Currency Prices with Deep Learning: Short to Long Horizon
DOI:
https://doi.org/10.70112/ajcst-2026.15.1.4416Keywords:
Machine Learning, Financial Economics, LSTM Neural Networks, Model Prediction, Currency ForecastingAbstract
The fast-evolving cryptocurrency markets present both special opportunities and challenges. The risk involved in investing in cryptocurrency assets is very high because exchange prices can change on a day-to-day basis. The study uses powerful machine learning techniques to forecast the value of cryptocurrencies. In comparison to the other seven models with fewer errors, neural networks achieved the best forecasting and validation performance. In order to predict future trends, LSTM (Long Short-Term Memory) neural networks were used. Complex relationships in financial data can be effectively analyzed using the LSTM model. Overall, more than fifty cryptocurrencies were subjected to Exploratory Data Analysis (EDA), which began with the collection of historical data and continued with feature engineering, integrative binning, data preparation, and standardization. The most successful ones were identified based on price movement, market size, and trading volume. The LSTM-based model was coded in Python and applied to 90-day price movement data to examine the existence of complex patterns and correlations. The performance indicators used to monitor the model were RMSE and MAE. These results corroborate the Adaptive Market Hypothesis (AMH), which posits that changes in investor and market behavior impact the dynamic efficiency of cryptocurrency markets. As shown in the paper, machine learning models have significant potential in financial economics and can be beneficial for investment decision-making processes and risk management approaches.
References
[1] Agyei, S. K., et al., "Does volatility in cryptocurrencies drive the interconnectedness between the cryptocurrencies market?" Cogent Economics & Finance, vol. 10, no. 1, p. 2061682, 2022.
[2] Akbulaev, N., et al., "Correlation and Regression Analysis of the Relation between Ethereum Price and Both Its Volume and Bitcoin Price," Journal of Structured Finance, vol. 26, no. 2, pp. 46–56, 2020.
[3] Awotunde, J. B., et al., Machine learning algorithm for cryptocurrencies price prediction. Springer, 2021.
[4] Borio, C., Furfine, C., and Lowe, P., "Procyclicality of the financial system and financial stability," BIS Papers, vol. 1, no. 3, pp. 1–57, 2001.
[5] Dos Santos, R. P., "On the philosophy of bitcoin/blockchain technology," Metaphilosophy, vol. 48, no. 5, pp. 620–633, 2017.
[6] Fang, F., et al., "Cryptocurrency trading: a comprehensive survey," Financial Innovation, vol. 8, no. 1, p. 13, 2022.
[7] Florencio, F., and Moreno, E. D., "Benchmarking the Keras API on GPU," International Journal of High Performance Computing and Networking, vol. 17, no. 1, pp. 19–27, 2021.
[8] Ghosh, I., Jana, R. K., and Sanyal, M. K., "Predictive modeling of financial markets using machine learning algorithms," Applied Soft Computing, vol. 82, p. 105553, 2019.
[9] Habib, G., et al., "Blockchain technology: benefits, challenges, applications," Future Internet, vol. 14, no. 11, p. 341, 2022.
[10] Huyen Chau, N. T., and Doan, T. P., Data Processing and Feature Engineering for Stock Price Trend Prediction. Springer, 2024.
[11] Jain, S., and Chauhan, D., Standard Multiple Regression Analysis Model, 2019.
[12] Kappal, S., "Data normalization using MMAD-based Z-score for robust predictions," London Journal of Research in Science, vol. 19, no. 4, pp. 39–44, 2019.
[13] Karn, P. K., et al., "Generalized Framework for Liquid Neural Network upon Sequential and Non-Sequential Tasks," Mathematics, vol. 12, no. 16, 2024.
[14] LeCun, Y., Bengio, Y., and Hinton, G., "Deep learning," Nature, vol. 521, no. 7553, pp. 436–444, 2015.
[15] Lechner, M., and Hasani, R., "Learning long-term dependencies in irregularly-sampled time series," arXiv preprint arXiv:2006.04418, 2020.
[16] Lindemann, B., et al., "A survey on long short-term memory networks for time series prediction," Procedia CIRP, vol. 99, pp. 650–655, 2021.
[17] Mudassir, M., et al., "Time-series forecasting of Bitcoin prices using high-dimensional features," Neural Computing and Applications, pp. 1–15, 2020.
[18] Naghib Moayed, A., and Habibi, R., "Crypto-Currency Price Prediction with Decision Tree Based Regressions," Journal of Algorithms and Computation, vol. 52, no. 2, pp. 29–40, 2020.
[19] Nakamoto, S., "Bitcoin: A peer-to-peer electronic cash system," Bitcoin.org, 2008.
[20] Oliva, G. A., et al., "An exploratory study of smart contracts in Ethereum," Empirical Software Engineering, vol. 25, pp. 1864–1904, 2020.
[21] Peng, Y., et al., "Feature selection and deep neural networks for stock price direction forecasting," Machine Learning with Applications, vol. 5, p. 100060, 2021.
[22] Taskinsoy, J., "Bitcoin: A New Digital Gold Standard in the 21st Century," SSRN 3941857, 2021.
[23] Truong, D., Data Science and Machine Learning for Non-Programmers: Using SAS Enterprise Miner. CRC Press, 2024.
[24] Urolagin, S., Sharma, N., and Datta, T. K., "Multivariate LSTM with Z-Score transformations for price forecasting," Energy, vol. 231, p. 120963, 2021.
[25] Zollanvari, A., Supervised Learning in Practice: Application Using Scikit-Learn. Springer, 2023.
[26] R. Kumar and S. Verma, "Deep learning approaches for financial time series forecasting," Asian Journal of Computer Science and Technology, vol. 11, no. 2, pp. 15–21, 2022.
[27] P. Reddy and M. Kumar, "Machine learning based prediction models for cryptocurrency markets," Asian Journal of Computer Science and Technology, vol. 12, no. 1, pp. 34–40, 2023.
[28] S. Rao and K. Prasad, "Comparative analysis of deep learning models for stock and crypto price prediction," CRI Journal of Technology and Computing, vol. 5, no. 1, pp. 22–29, 2022.
[29] A. Sharma and R. Gupta, "LSTM based financial forecasting using big data analytics," CRI Journal of Computer and Information Sciences, vol. 6, no. 2, pp. 41–48, 2023.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Centre for Research and Innovation
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
