Performance Comparison of GRU and LSTM Methods for Predicting Bitcoin Exchange Rate against US Dollar
(1) * Ronal Kepo Mail (Papua New Guinea University of Technology, Papua New Guinea)
(2) Daniel Okokpujie Mail (Papua New Guinea University of Technology, Papua New Guinea)
*corresponding author

Abstract


This research aims to compare the performance between the Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) methods in predicting the Bitcoin exchange rate against the US Dollar (BTC-USD). The data used comes from Yahoo Finance for the period 2017-2022. Each model is built with a comparable architecture and evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), coefficient of determination (R²), and prediction accuracy metrics. The results show that the LSTM model performed better on the test data with a MAPE of 3.80% and an accuracy of 96.20%, while the GRU model achieved a MAPE of 5.13% and an accuracy of 94.87%. Although the GRU model performed better on the training data, the LSTM model showed better generalization ability on the testing data. This research provides important insights into the selection of the optimal recurrent neural network architecture for Bitcoin exchange rate prediction which is known for its high volatility.

   

DOI

https://doi.org/10.33292/ijarlit.v2i1.32 		      

Article metrics

10.33292/ijarlit.v2i1.32 Abstract views : 121 | PDF views : 58

   

Cite

How to cite item
   

Full Text

Download

References


A. F. Bariviera, "The inefficiency of Bitcoin revisited: A dynamic approach," Econ. Lett., vol. 161, pp. 1-4, 2017.

M. Mudassir, S. Bennbaia, D. Unal, and M. Hammoudeh, "Time-series forecasting of Bitcoin prices using high-dimensional features: a machine learning approach," Neural Comput. Appl., pp. 1-15, 2020.

S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Comput., vol. 9, no. 8, pp. 1735-1780, 1997.

K. Cho et al., "Learning phrase representations using RNN encoder-decoder for statistical machine translation," arXiv Prepr. arXiv1406.1078, 2014.

K. He and J. Qian, "Research on Stock Prediction Algorithm Based on CNN and LSTM," Acad. J. Comput. Inf. Sci., vol. 5, no. 12, 2022, doi: 10.25236/ajcis.2022.051215.

F. Wang, D. Zhang, G. Min, and J. Li, "Reservoir Production Prediction Based on Variational Mode Decomposition and Gated Recurrent Unit Networks," Ieee Access, vol. 9, pp. 53317-53325, 2021, doi: 10.1109/access.2021.3070343.

Q. Fang and X. Maldague, "A Method of Defect Depth Estimation for Simulated Infrared Thermography Data With Deep Learning," Appl. Opt. Sci., vol. 10, no. 19, p. 6819, 2020, doi: 10.3390/app10196819.

N. Elsayed, S. Anthony, and M. Bayoumi, "Deep Gated Recurrent and Convolutional Network Hybrid Model for Univariate Time Series Classification," Int. J. Adv. Comput. Sci. Appl., vol. 10, no. 5, 2019, doi: 10.14569/ijacsa.2019.0100582.

C. S. Alfredo and D. Adytia, "Time Series Forecasting of Significant Wave Height Using GRU, CNN-GRU, and LSTM," J. Resti (Information Systems Engineering and Technol. Information), vol. 6, no. 5, pp. 776-781, 2022, doi: 10.29207/resti.v6i5.4160.

D. Zhang, Q. Peng, J. Lin, D. Wang, X. Liu, and J. Zhuang, "Simulating Reservoir Operation Using a Recurrent Neural Network Algorithm," Water, vol. 11, no. 4, p. 865, 2019, doi: 10.3390/w11040865.

H. M. Lynn, S. B. Pan, and P. Kim, "A Deep Bidirectional GRU Network Model for Biometric Electrocardiogram Classification Based on Recurrent Neural Networks," Ieee Access, vol. 7, pp. 145395-145405, 2019, doi: 10.1109/access.2019.2939947.

T. Fischer and C. Krauss, "Deep learning with long short-term memory networks for financial market predictions," Eur. J. Oper. Res., vol. 270, no. 2, pp. 654-669, 2018.

H. Y. Kim and C. H. Won, "Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models," Expert Syst. Appl., vol. 103, pp. 25-37, 2018.

S. Ahmed, S. Mubarak, J. T. Du, and S. Wibowo, "Forecasting the Status of Municipal Waste in Smart Bins Using Deep Learning," Int. J. Environ. Res. Public Health, vol. 19, no. 24, p. 16798, 2022, doi: 10.3390/ijerph192416798.

Z. Halim, S. M. Shuhidan, and Z. M. Sanusi, "Corporation Financial Distress Prediction With Deep Learning: Analysis of Public Listed Companies in Malaysia," Bus. Process Manag. J., vol. 27, no. 4, pp. 1163-1178, 2021, doi: 10.1108/bpmj-06-2020-0273.

H. ÇET?NER, "Recurrent Neural Network Based Model Development for Energy Consumption Forecasting," Bitlis Eren Üniversitesi Fen Bilim. Derg., vol. 11, no. 3, pp. 759-769, 2022, doi: 10.17798/bitlisfen.1077393.

D. Chicco, M. J. Warrens, and G. Jurman, "The Coefficient of Determination R-Squared Is More Informative Than SMAPE, MAE, MAPE, MSE and RMSE in Regression Analysis Evaluation," Peerj Comput. Sci., vol. 7, p. e623, 2021, doi: 10.7717/peerj-cs.623.

Y. Li, Z. Zhu, D. Kong, H. Han, and Y. Zhao, "EA-LSTM: Evolutionary Attention-Based LSTM for Time Series Prediction," Knowledge-Based Syst., vol. 181, p. 104785, 2019, doi: 10.1016/j.knosys.2019.05.028.

P. Xiu-li, Q. Li, W. Yannian, and Y. Deng-feng, "High-Precision Blood Glucose Prediction and Hypoglycemia Warning Based on the LSTM-GRU Model," Trends Comput. Sci. Inf. Technol., vol. 7, no. 3, pp. 74-80, 2022, doi: 10.17352/tcsit.000053.

A. K. Singh et al., "A global review of rubber plantations: Impacts on ecosystem functions, mitigations, future directions, and policies for sustainable cultivation," Sci. Total Environ., vol. 796, p. 148948, Nov. 2021, doi: 10.1016/j.scitotenv.2021.148948.


Refbacks

  • There are currently no refbacks.


View My Stats IJARLIT