Document Type : Research Article

Author

Department of Computer Science, Shaqra University, Shaqraa, Saudi Arabia

Abstract

The increasing volatility in pricing and growing potential for profit in digital currency have made predicting the price of cryptocurrency a very attractive research topic. Several studies have already been conducted using various machine-learning models to predict crypto currency prices. This study presented in this paper applied a classic Autoregressive Integrated Moving Average(ARIMA) model to predict the prices of the three major cryptocurrencies âAT Bitcoin, XRP and Ethereum âAT using daily, weekly and monthly time series. The results demonstrated that ARIMA outperforms most other methods in predicting cryptocurrency prices on a daily time series basis in terms of mean absolute error (MAE), mean squared error (MSE) and root mean squared error(RMSE).

Keywords

[1] Noorhazlina Abu Bakar and Sofian Rosbi. Autoregressive integrated moving average ( arima ) model for forecasting cryptocurrency exchange rate in high volatility environment : A new insight of bitcoin transaction. 2017.
[2] Jonathan Rebane, Isak Karlsson, Panagiotis Papapetrou, and Stojan Denic. Seq2seq rnns and arima models for cryptocurrency prediction : A comparative study. In Proceedings of SIGKDD Workshop on Fintech (SIGKDD FintechâAZ18)
:, 2018.
[3] T Guo and N Antulov-Fantulin. An experimental study of bitcoin fluctuation using machine learning methods. arXiv preprint arXiv:1802.04065,2018.
[4] Sean McNally, Jason Roche, and Simon Caton. Predicting the price of bitcoin using machine  learning. In 2018 26th Euromicro International Conference on Parallel, Distributed and Networkbased Processing (PDP), pages 339–343. IEEE,
2018.
[5] Alex Greaves and Benjamin Au. Using the bitcoin transaction graph to predict the price of bitcoin. No Data, 2015.
[6] Huisu Jang and Jaewook Lee. An empirical study on modeling and prediction of bitcoin prices with bayesian neural networks based on blockchain information. IEEE Access, 6:5427–5437, 2018.
[7] Thearasak Phaladisailoed and Thanisa Numnonda. Machine learning models comparison for bitcoin price prediction. In 2018 10th International Conference on Information Technology and Electrical Engineering (ICITEE), pages 506–511. IEEE, 2018.
[8] Yaohao Peng, Pedro Henrique Melo Albuquerque, Jader Martins Camboim de Sá, Ana Julia Akaishi Padula, and Mariana Rosa Montenegro. The best of two worlds: Forecasting high frequency volatility for cryptocurrencies and traditional
currencies with support vector regression. Expert Systems with Applications, 97:177–192,2018.
[9] Kaggle. http:// www.kaggle.com. accessed 15 June 2018.
[10] Coinmarktcap. http:// www. coinmarketcap.com. accessed 18 Dec 2018.
[11] Isaac Madan, Shaurya Saluja, and Aojia Zhao. Automated bitcoin trading via machine learning algorithms. 2014.
[12] Javier Contreras, Rosario Espinola, Francisco J Nogales, and Antonio J Conejo. Arima models to predict next-day electricity prices. IEEE transactions on power systems, 18(3):1014–1020,2003.
[13] Dennys CA Mallqui and Ricardo AS Fernandes. Predicting the direction, maximum, minimum and closing prices of daily bitcoin exchange rate using machine learning techniques. Applied Soft Computing, 75:596–606, 2019.
[14] Siddhi Velankar, Sakshi Valecha, and Shreya Maji. Bitcoin price prediction using machine learning. In 2018 20th International Conference on Advanced Communication Technology(ICACT), pages 144–147. IEEE, 2018.
[15] Christos Agiakloglou and Paul Newbold. Empirical evidence on dickey-fuller-type tests. Journal of Time Series Analysis, 13(6):471–483, 1992.