Hyperparameter Tuning of the Prophet Model Using Particle Swarm Optimization: A Case Study on Ethereum: Hyperparameter Tuning of the Prophet Model Using Particle Swarm Optimization

Kwestan Ahmed Ismael; Heshu Othman Faqe; Mohammed Hussein Abdalla; Hindreen A. Taher

doi:10.63841/31651

Authors

Kwestan Ahmed Ismael Department of Mathematics, College of Basic Education, University of Raparin, Sulaimani, IRAQ Author https://orcid.org/0009-0001-4562-7961
Heshu Othman Faqe Department of Statistics and Informatics, College of Administration and Economic, University of Sulaimani, Sulaimani, IRAQ Author https://orcid.org/0000-0002-4494-0948
Mohammed Hussein Abdalla Department of Computer Science, College of Basic Education, University of Raparin, Sulaimani, IRAQ Author https://orcid.org/0000-0003-0973-8389
Hindreen A. Taher Department of Information Technology, College of Commerce, University of Sulaimani, Sulaimani, IRAQ;Department of Software Engineering, Faculty of Engineering & Computer Science, Qaiwan International University Sulaymaniyah, Kurdistan Region-IRAQ Author https://orcid.org/0000-0002-8405-8570

DOI:

https://doi.org/10.63841/31651

Keywords:

Ethereum, Prophet model, Particle Swarm Optimizer, Seasonality, Time series.

Abstract

In this work we use historical market data from Bitget to predict weekly open prices of Ethereum (ETH) for a 96-week period with the Prophet forecast model trained by using Particle Swarm Optimization (PSO) algorithm. Because of this, the research delves into automated hyperparameter tuning for Prophet in order to improve forecast performance on cryptocurrency markets where volatility, structural breaks and irregular trading patterns pose a significant challenge to time series prediction. The PSO algorithm is a good method to explore the high dimensional parameter space in which it can strike between the global analysis and local exploitation for detecting minimal forecast errors. Based on evaluating model performance for which we used accuracy metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) in training, test holdout & full-fit situations. PSO-optimized Prophet: The results show a great in-sample fitting and fast convergence behaviour, as the best CV RMSE is slightly higher than the lowest one should have obtained if used only 10 iterations. Although forecasts exhibit stability and track long-term trends well, the model does not predict short-term fluctuations in the holdout set with high accuracy (wider forecast uncertainty intervals). Our results shed light on the utility of PSO to improve Prophet-based price prediction in cryptocurrencies, reinforce the relevance of uncertainty quantification in asset markets and inform risk-aware decisions of financial agents dealing with unstable assets.

Downloads

Download data is not yet available.

Author Biographies

Kwestan Ahmed Ismael, Department of Mathematics, College of Basic Education, University of Raparin, Sulaimani, IRAQ

Kwestan Ahmed Ismael is an Assistant Lecturer at the Department of Mathematics, College of Basic Education, University of Raparin. She got the B.Sc. degree in Statistical Sciences from the College of Administration and Economics, University of Sulaimani, in 2009–2010, and the M.Sc. degree in Statistics from the University of Sulaimani in 2015–2016. Her research interests are in statistical analysis, data modeling, and applied mathematics.
Heshu Othman Faqe, Department of Statistics and Informatics, College of Administration and Economic, University of Sulaimani, Sulaimani, IRAQ

Heshu Othman Faqe is a Lecturer at the Department of Statistics and Informatics, College of Administration and Economic University of Sulaimani. She got the B.Sc. degree in Statistics Science from the College of Administration and Economics, University of Sulaimani, in 2006–2007, and the M.Sc. degree in Statistics Science from the College of Administration and Economics, University of Sulaimani, in 2018. Her research interests are in statistical analysis, linear modeling, and applied statistics.
Mohammed Hussein Abdalla, Department of Computer Science, College of Basic Education, University of Raparin, Sulaimani, IRAQ

Mohammed Hussein Abdalla Meera is an Assistant Lecturer at the Department of computer, College of Basic Education, University of Raparin. He got the B.Sc. degree in computer Sciences from the College of Sciences, University of Sulaimani, in 2011–2012, and the M.Sc. degree in computer Sciences from the University of Firat/ Turkey in 2020–2016. His research interests are in statistical analysis, data modeling, and applied mathematics.
Hindreen A. Taher, Department of Information Technology, College of Commerce, University of Sulaimani, Sulaimani, IRAQ;Department of Software Engineering, Faculty of Engineering & Computer Science, Qaiwan International University Sulaymaniyah, Kurdistan Region-IRAQ

Hindreen A. Taher is an Assistant Professor at the Department of Information and Technology, College of Commerce, University of Sulaimani. He got the B.Sc. degree in Statistics Science from the College of Administration and Economic, University of Sulaimani, in 2010–2011, and the M.Sc. degree in Statistics Science from the College of Administration and Economic, University of Sulaimani, in 2016, and PhD in Bayesian Statistics and Bayesian Artificial Intelligence (AI) from the College of Administration and Economic, University of Sulaimani, in 2023 His research interests are in statistical analysis, data modeling, machine learning, deep learning and applied mathematics

References

Y. Chen and C. Bellavitis, “Blockchain disruption and decentralized finance: The rise of decentralized business models,” J. Bus. Venturing Insights, vol. 13, p. e00151, 2020.

D. Shen, A. Urquhart, and P. Wang, “Forecasting the volatility of Bitcoin: The importance of jumps and structural breaks,” Eur. Financ. Manag., vol. 25, no. 2, pp. 289–315, 2019.

J. Kennedy and R. Eberhart, “Particle swarm optimization,” in Proc. ICNN’95 - Int. Conf. Neural Netw., vol. 4, pp. 1942–1948, 1995.

S. J. Taylor and B. Letham, “Forecasting at scale,” Amer. Statist., vol. 72, no. 1, pp. 37–45, 2018.

J. Ma and E. Mahmoudinia, “Comprehensive modeling approaches for forecasting Bitcoin transaction fees: A comparative study,” arXiv preprint arXiv:2502.01029, 2025. [Online]. Available: https://arxiv.org/abs/2502.01029.

S.-H. Choi, S.-M. Choi, and S.-J. Buu, “Proximal policy-guided hyperparameter optimization for mitigating model decay in cryptocurrency scam detection,” Electronics, vol. 14, no. 6, p. 1192, 2025. [Online]. Available: https://www.mdpi.com/2079-9292/14/6/1192.

A. A. Aziz, B. M. Shafeeq, R. A. Ahmed, and H. A. Taher, “Employing recurrent neural networks to forecast the dollar exchange rate in the parallel market of Iraq,” Tikrit J. Admin. Econ. Sci., vol. 19, no. 62, p. 2, 2023.

Y. Shi and R. C. Eberhart, “A modified particle swarm optimizer,” in Proc. IEEE Int. Conf. Evol. Comput., pp. 69–73, 1998.

R. C. Eberhart and Y. Shi, “Particle swarm optimization: Developments, applications and resources,” in Proc. Congr. Evol. Comput., vol. 1, pp. 81–86, 2001.

M. Clerc and J. Kennedy, “The particle swarm—explosion, stability, and convergence in a multidimensional complex space,” IEEE Trans. Evol. Comput., vol. 6, no. 1, pp. 58–73, 2002.

A. Ratnaweera, S. K. Halgamuge, and H. C. Watson, “Self-organizing hierarchical particle swarm optimizer with time-varying acceleration coefficients,” IEEE Trans. Evol. Comput., vol. 8, no. 3, pp. 240–255, 2004.

R. Poli, J. Kennedy, and T. Blackwell, “Particle swarm optimization: An overview,” Swarm Intell., vol. 1, pp. 33–57, 2007.

M. Zambrano-Bigiarini, M. Clerc, and R. Rojas, “Standard particle swarm optimisation 2011 at CEC-2013: A baseline for future PSO improvements,” in Proc. IEEE CEC, pp. 2337–2344, 2013.

A. Asraa, W. Rodeen, and H. Tahir, “Forecasting the impact of waste on environmental pollution,” Int. J. Sustain. Develop. Sci., vol. 1, no. 1, pp. 1–12, 2018.

B. K. Ahmed, S. A. Rahim, B. B. Maaroof, and H. A. Taher, “Comparison between ARIMA and Fourier ARIMA model to forecast the demand of electricity in Sulaimani Governorate,” Qalaai Zanist J., vol. 5, no. 3, pp. 908–940, 2020.

S. T. Ahmed and H. A. Tahir, “Comparison between GM (1, 1) and FOGM (1, 1) models for forecasting the rate of precipitation in Sulaimani,” J. Kirkuk Univ. Admin. Econ. Sci., vol. 11, no. 1, pp. 301–314, 2021.

Hyperparameter Tuning of the Prophet Model Using Particle Swarm Optimization: A Case Study on Ethereum