AN ARIMA-NARNN HYBRID TIME SERIES ANALYSIS OF DAILY BITCOIN PRICE DURING AND AFTER THE COVID-19 PANDEMIC OUTBREAK

Catherine Ngo; Orson Chi; Yeong Nain Chi

doi:10.29121/shodhai.v2.i2.2025.45

Authors

Catherine Ngo Department of Agriculture, Food, and Resource Sciences, University of Maryland Eastern Shore Princess Anne, MD 21853 U.S.A
Orson Chi Department of Computer Science and Engineering Technology, University of Maryland Eastern Shore, Princess Anne, MD 21853 U.S.A
Yeong Nain Chi Department of Agriculture, Food, and Resource Sciences, University of Maryland Eastern Shore Princess Anne, MD 21853 U.S.A.

DOI:

https://doi.org/10.29121/shodhai.v2.i2.2025.45

Keywords:

Bitcoin, Price, Time Series, Covid-19, Arima Model, Narnn Model, Arima-Narnn Hybrid Model

Abstract

This study explored the suitability of an advanced hybrid model to predict the very volatile Bitcoin prices during and after the COVID-19 pandemic. At this crucial point, the Bitcoin trend saw a steep rise and exhibited significant fluctuations. Therefore, the effectiveness of the traditional ARIMA and ARIMA-NARNN hybrid models was tested and compared. Following the Box–Jenkins methodology, the ARIMA(0,1,1) with drift model was identified as the best-fit model for the time series because of its lowest AIC value. While ARIMA models excel in modeling linear problems within time-series data, NARNN models are better suited for nonlinear patterns. However, an ARIMA-NARNN hybrid model was explored, which combines the strengths of both ARIMA and NARNN models, offering the capability to address both linear and nonlinear aspects of time series data. The comparative analysis of this study demonstrated that the ARIMA-NARNN hybrid model, with 10 neurons in the hidden layer and 2 time delays, outperformed the ARIMA(0,1,1) model with the lowest MSE. These findings represent a significant step in time series forecasting by leveraging the strengths of both statistical and ML methods.

References

Atsalakis, G. S., Atsalaki, I. G., Pasiouras, F., & Zopounidis, C. (2019). Bitcoin Price Forecasting with Neuro-Fuzzy Techniques. European Journal of Operational Research, 276(2), 770–780. https://doi.org/10.1016/j.ejor.2019.01.040

Ayaz, Z., Fiajdhi, J., Sabah, A., & Ansari, M. (2020). Bitcoin Price Prediction Using ARIMA Model. TechRxiv, 1–10. https://doi.org/10.36227/techrxiv.12098067.v1

Boozary, P., Sheykhan, S., & GhorbanTanhaei, H. (2025). Forecasting the Bitcoin Price Using the Various Machine Learning: A Systematic Review in Data-Driven Marketing. Systems and Soft Computing, 7, Article 200209, 1–13. https://doi.org/10.1016/j.sasc.2025.200209

Denton, J. W. (1995). How Good are Neural Networks for Causal Forecasting. The Journal of Business Forecasting Methods & Systems, 14, 17.

Dowe, D. L., Peiris, S., & Kima, E. (2025). A Novel ARFIMA-ANN Hybrid Model for Forecasting Time Series—and its Role in Explainable AI. Journal of Econometrics and Statistics, 5(1), 107–127.

Fathi, O. (2025). Time Series Forecasting Using a Hybrid ARIMA and LSTM Model. Velvet Consulting.

Hajirahimi, Z., & Khashei, M. (2023). Hybridization of Hybrid Structures for Time Series Forecasting: A Review. Artificial Intelligence Review, 56, 1201–1261. https://doi.org/10.1007/s10462-022-10199-0

Hao, Y., & Gao, Q. (2020). Predicting the Trend of Stock Market Index Using the Hybrid Neural Network Based on Multiple Time Scale Feature Learning. Applied Sciences, 10(11), 3961. https://doi.org/10.3390/app10113961

Hozjan, V. (2025, February 25). Bitcoin: From birth to Global Recognition. NiceHash.

Laurent, D. (2025, February 25). The History of Bitcoin: From Inception to Mainstream Adoption. WoolyBlog.

Levenberg, K. (1944). A Method for the Solution of Certain Non-Linear Problems in Least Squares. Quarterly of Applied Mathematics, 2(2), 164–168. https://doi.org/10.1090/qam/10666

Luc Minh, T., Senkerik, R., & Dang, T. K. (2024). Predicting Bitcoin's Price: A Critical Review of Forecasting Models and Methods. In T. K. Dang, J. Küng, & T. M. Chung (Eds.), Future Data and Security Engineering. Big Data, Security and Privacy, Smart City and Industry 4.0 Applications FDSE 2024 (Vol. 2309). Springer. https://doi.org/10.1007/978-981-96-0434-0_3

Marquardt, D. W. (1963). An Algorithm for Least-Squares Estimation of Nonlinear Parameters. Journal of the Society for Industrial and Applied Mathematics, 11(2), 431–441. https://doi.org/10.1137/0111030

Ponciano, J. (2025, February 25). Tesla's Bitcoin Investment fell $1 billion last quarter amid crypto market crash. Forbes.

Portfolio Insider. (2025, February 25). Can Bitcoin Grow Faster than the Internet? Nasdaq.

Prajapati, S., Swaraj, A., Lalwani, R., Narwal, A. A., & Verma, K. (2021). Comparison of Traditional and Hybrid Time Series Models for Forecasting COVID-19 cases. arXiv, 1–14. https://doi.org/10.21203/rs.3.rs-493195/v1

Russolillo, S. (2025, February 25). Bitcoin Goes to the Big Four: PWC Accepts First Digital-Currency Payment. The Wall Street Journal.

Yenidogan, I., Cayır, A., Kozan, O., Dag, T., & Arslan, C. (2018). Bitcoin Forecasting Using ARIMA and PROPHET. In Proceedings of the 3rd International Conference on Computer Science and Engineering (UBMK), Bosnia. https://doi.org/10.1109/UBMK.2018.8566476

Yu, L., Zhou, L., Tan, L., Jiang, H., Wang, Y., Wei, S., & Nie, S. (2014). Application of a New Hybrid Model with Seasonal Auto-Regressive Integrated Moving Average (ARIMA) and Nonlinear Auto-Regressive Neural Network (NARNN) in Forecasting Incidence Cases of HFMD in Shenzhen, China. PLOS One, 9(6), e98241. https://doi.org/10.1371/journal.pone.0098241

Zhang, G. P. (2003). Time Series Forecasting Using a Hybrid ARIMA and Neural Network Model. Neurocomputing, 50, 159–175. https://doi.org/10.1016/S0925-2312(01)00702-0