Author
Listed:
- Badr Alnssyan
- Muhammad Ali
Abstract
Due to the nonlinear, nonstationary, and chaotic nature of the cryptocurrency prices, precise prediction still remains a challenge for both researchers and investors with direct implications for risk management, strategic investment, and algorithmic trading. In this study an attempt has been made to forecast the future daily closing prices of Bitcoin, Ethereum, XRP, and Binance Coin using 14 different models, of which six are classical statistical, namely, autoregressive integrated moving average (ARIMA), error trend seasonal (ETS), TBATS, Theta, Naïve, and Prophet and eight deep learning and hybrid architectures such as long-short term memory network (LSTM), recurrent neural network (RNN), convolution neural network (CNN), artificial neural network (ANN), temporal convolutional network (TCN), CNN-LSTM, BiLSTM-Attention, and LSTM-Autoencoder. The length of the data is from January 1, 2021, to August 14, 2026. All models are evaluated using a consistent chronological 80:20 train-test split, and to assess the forecast accuracy of these models, four performance measures, such as root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and the coefficient of determination (R2) are utilized. Furthermore, statistical significance was assessed using the Diebold–Mariano (DM) test, with the naïve random walk model serving as the benchmark model. The empirical findings show that, across all cryptocurrencies, the deep learning and hybrid models demonstrated statistically significant forecasting superiority over the benchmark naïve model. However, no single deep learning or hybrid architecture dominates uniformly, with model performance varying by asset, and hybrid architectures do not necessarily outperform simpler standalone models. These findings offer practical guidance for model selection in cryptocurrency price forecasting applications.
Suggested Citation
Badr Alnssyan & Muhammad Ali, 2026.
"Toward Robust Cryptocurrency Price Forecasting: A Comparative Analysis of Statistical, Deep Learning, and Hybrid Architectures,"
Complexity, Hindawi, vol. 2026, pages 1-21, November.
Handle:
RePEc:hin:complx:8631926
DOI: 10.1155/cplx/8631926
Download full text from publisher
Corrections
All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:hin:complx:8631926. See general information about how to correct material in RePEc.
If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.
We have no bibliographic references for this item. You can help adding them by using this form .
If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Mohamed Abdelhakeem (email available below). General contact details of provider: https://www.hindawi.com .
Please note that corrections may take a couple of weeks to filter through
the various RePEc services.