Closing Price Prediction of Cryptocurrencies BTC, LTC, and ETH Using a Hybrid ARIMA-LSTM Algorithm
Downloads
This study aims to develop a hybrid algorithm using the ARIMA model and LSTM-type recurrent neural networks to predict the closing prices of the cryptocurrencies BTC, LTC, and ETH. The methodology includes an exploratory data analysis, followed by the design, implementation, and evaluation of each individual algorithm as well as the combined hybrid algorithm. The results, after experimentation and evaluation of metrics on the test set, indicated that the ARIMA model was inefficient in predicting the closing prices of cryptocurrencies. On the other hand, the hybrid model for BTC showed significant statistical differences in the metrics, with MAE = $726.21 and MAPE = 1.75%, compared to the LSTM model, which achieved MAE = $729.35 and MAPE = 1.76%. These results indicate better performance from the hybrid model. Regarding the RMSE metric, the hybrid model scored 1157.47, while LSTM scored 1159.99; although statistically equivalent, the hybrid model was numerically better. For the remaining metrics and other cryptocurrencies, both methods were statistically equivalent. For five-day-ahead predictions, the hybrid algorithm continued to yield better results for LTC and ETH.
Downloads
[1] Acosta Valderrama, J. C. (2018). SABER UCV: Study of Cryptocurrencies Using Time Series and Neural Networks. Universidad Central de Venezuela, Caracas, Venezuela.
[2] E.F.P.A. (2024). Financial advisors, their great challenges. Asesores Financieros, EFPA (Asociación Española de Asesores y Planificadores Financiero), 15, 18–20. Available online: https://www.asesoresfinancierosefpa.es/ejemplares-revista-asesores-financieros-efpa/los-grandes-desafios-del-sector-en-portada-del-numero-15-de-asesores-financieros-efpa/ (accessed on May 2025).
[3] López-García, A. (2023). Comparative analysis of deep neural networks for Bitcoin market price prediction. Revista Electrónica de Comunicaciones y Trabajos de ASEPUMA, 24(1), 1–16. doi:10.24310/recta.24.1.2023.19858.
[4] Azari, A. (2019). Bitcoin price prediction: An ARIMA approach. arXiv preprint arXiv:1904.05315, 1-3. doi:10.48550/arXiv.1904.05315.
[5] Yu, S. (2024). Comparative analysis of machine learning techniques for cryptocurrency price prediction. Applied and Computational Engineering, 32(1), 1–12. doi:10.54254/2755-2721/32/20230175.
[6] Jiménez Gómez, L. M., Lambis-Alandete, E., & Velásquez-Henao, J. D. (2023). Comparison of Deep Learning Algorithms for Cryptocurrency Price Forecasting. Ingeniería y Competitividad, 25(3). doi:10.25100/iyc.v25i3.12845.
[7] Siami-Namini, S., Tavakoli, N., & Namin, A. S. (2019). The Performance of LSTM and BiLSTM in Forecasting Time Series. Proceedings: IEEE International Conference on Big Data, Big Data 2019, 3285–3292. doi:10.1109/BigData47090.2019.9005997.
[8] Xu, D., Zhang, Q., Ding, Y., & Zhang, D. (2022). Application of a hybrid ARIMA-LSTM model based on the SPEI for drought forecasting. Environmental Science and Pollution Research, 29(3), 4128–4144. doi:10.1007/s11356-021-15325-z.
[9] Pierre, A. A., Akim, S. A., Semenyo, A. K., & Babiga, B. (2023). Peak Electrical Energy Consumption Prediction by ARIMA, LSTM, GRU, ARIMA-LSTM and ARIMA-GRU Approaches. Energies, 16(12), 4739. doi:10.3390/en16124739.
[10] Bouteska, A., Abedin, M. Z., Hajek, P., & Yuan, K. (2024). Cryptocurrency price forecasting – A comparative analysis of ensemble learning and deep learning methods. International Review of Financial Analysis, 92, 1–12. doi:10.1016/j.irfa.2023.103055.
[11] Flores-Fernandez, G. A., Jimenez-Carrion, M., Gutierrez, F., & Sanchez-Ancajima, R. A. (2024). Genetic Algorithm and LSTM Artificial Neural Network for Investment Portfolio Optimization. Journal of Wireless Mobile Networks, Ubiquitous Computing, and Dependable Applications, 15(2), 27–46. doi:10.58346/JOWUA.2024.I2.003.
[12] Cuthbertson, A. (2020). Bitcoin price reaches its highest value in history. Independentespanol. Available online: https://www.independentespanol.com/tecnologia/bitcoin-precio-valor-mas-alto-historia-b1764093.html (accessed on May 2025).
[13] Astuti, I. D., Rajab, S., & Setiyouji, D. (2022). Cryptocurrency Blockchain Technology in the Digital Revolution Era. APTISI Transactions on Technopreneurship, 4(1), 9–15. doi:10.34306/att.v4i1.216.
[14] Darda, S. D. (2023). Advantages & Disadvantages of Cryptography & Valuation Analysis. National High School Journal of Science, 1–11.
[15] Mangiwa, R. D. D., Siregar, R., Sari, S. D., & Agustina, N. (2025). Application of the Arima Model in Forecasting Ethereum Prices. Parameter: Jurnal Matematika, Statistika Dan Terapannya, 4(1), 81–94. doi:10.30598/parameterv4i1pp81-94.
[16] Zhang, X. (2024). Analyzing Financial Market Trends in Cryptocurrency and Stock Prices Using CNN-LSTM Models. Preprints, 1-12. doi:10.20944/preprints202407.1119.v1.
[17] Tumpa, S. N., & Maduranga, K. D. G. (2024). Utilizing RNN for real-time cryptocurrency price prediction and trading strategy optimization. arXiv preprint arXiv:2411.05829. doi:10.48550/arXiv.2411.05829.
[18] Pirkhedri, A. (2025). Predicting Solana Cryptocurrency Prices in 2024: A Comparative Study of LSTM and GRU Models. International Journal of Industrial Engineering and Operational Research, 7(2), 11-19..
[19] Kaur, R., Uppal, M., Gupta, D., Juneja, S., Arafat, S. Y., Rashid, J., Kim, J., & Alroobaea, R. (2025). Development of a cryptocurrency price prediction model: leveraging GRU and LSTM for Bitcoin, Litecoin and Ethereum. PeerJ Computer Science, 11, 2675. doi:10.7717/peerj-cs.2675.
[20] Kabo, I. G., Obunadike, G. N., & Samaila, N. A. (2025). Sentiment-Driven and Economic Indicators for Bitcoin Price Forecasting: A Hybrid Time Series Model. Journal of Science Research and Reviews, 2(1), 99–107. doi:10.70882/josrar.2025.v2i1.34.
[21] Fan, D., Sun, H., Yao, J., Zhang, K., Yan, X., & Sun, Z. (2021). Well production forecasting based on ARIMA-LSTM model considering manual operations. Energy, 220, 119708. doi:10.1016/j.energy.2020.119708.
[22] Dave, E., Leonardo, A., Jeanice, M., & Hanafiah, N. (2021). Forecasting Indonesia Exports using a Hybrid Model ARIMA-LSTM. Procedia Computer Science, 179, 480–487. doi:10.1016/j.procs.2021.01.031.
[23] Gutiérrez, H., & De la Vara, R. (2008). Análisis y diseño de experimentos. México DF: McGraw-Hill Interamericana. Available online: https://gc.scalahed.com/recursos/files/r161r/w19537w/analisis_y_diseno_experimentos.pdf (accessed on May 2025).
[24] Klimberg, R. K., Sillup, G. P., Boyle, K. J., & Tavva, V. (2010). Forecasting performance measures - What are their practical meaning? Advances in Business and Management Forecasting, 7, 137–147. doi:10.1108/S1477-4070(2010)0000007012.
[25] Fisher, A., Rudin, C., & Dominici, F. (2019). All models are wrong, but many are useful: Learning a variable’s importance by studying an entire class of prediction models simultaneously. Journal of Machine Learning Research, 20, 1–81.
[26] Regal, A., Morzán, J., Fabbri, C., Herrera, G., Yaulli, G., Palomino, A., & Gil, C. (2019). Cryptocurrency price projection based on tweets using LSTM. Ingeniare. Revista Chilena de Ingeniería, 27(4), 696–706. doi:10.4067/s0718-33052019000400696.
[27] Jorge Dagnino, S. (2014). Datos faltantes (missing values). Revista Chilena de Anestesia, 43(4), 332–334.
[28] Nuñez Sánchez, N. D. (2023). Análisis de algoritmos para la imputación de datos y modelos de predicción: Informe final de pasantía para optar por el título de Matemática. Universidad Distrital Francisco José de Caldas. Available online: https://repository.udistrital.edu.co/server/api/core/bitstreams/d84b225f-429e-4517-9481-0a416f41d0a9/content (accessed on May 2025).
[29] Arranz Barcenilla, V. (2022). Aplicación web para la gestión de carteras de inversión usando técnicas de Inteligencia Artificial (Final Degree Project, Universidad de Valladolid, Escuela de Ingeniería Informática). UVaDOC. Available online: https://uvadoc.uva.es/handle/10324/57220 (accessed on May 2025).
- This work (including HTML and PDF Files) is licensed under a Creative Commons Attribution 4.0 International License.
