Time Series Forecasting Based on Deep Learning CNN-LSTM-GRU Model on Stock Prices

Time Series Forecasting Based on Deep Learning CNN-LSTM-GRU Model on Stock Prices
  IJETT-book-cover           
  
© 2023 by IJETT Journal
Volume-71 Issue-6
Year of Publication : 2023
Author : Ghani Rizky Naufal, Antoni Wibowo
DOI : 10.14445/22315381/IJETT-V71I6P215

How to Cite?

Ghani Rizky Naufal, Antoni Wibowo, "Time Series Forecasting Based on Deep Learning CNN-LSTM-GRU Model on Stock Prices," International Journal of Engineering Trends and Technology, vol. 71, no. 6, pp. 126-133, 2023. Crossref, https://doi.org/10.14445/22315381/IJETT-V71I6P215

Abstract
An equity, commonly referred to as a stock, is a financial asset representing partial ownership in the company that issued it. Stocks are favored among investors due to their potential for quick financial gains. It is important to note that stock prices are not arbitrary but instead follow specific patterns and can be analyzed and predicted using discrete time series models. This allows for studying and forecasting stock prices, enabling investors to make informed decisions. Much research has been done to forecast stock prices. The Long Short-Term Memory (LSTM) model, known for its popularity in deep learning, lacks the necessary strength to serve as the main approach for time series forecasting. To address the drawbacks, many various elements employing many methodologies and procedures are addressed; one option is to integrate several deep learning models to create a hybrid deep learning model. Researchers have utilized hybrid deep learning models LSTM-GRU and CNN-LSTM to forecast and analyze time series data, and both models outperform LSTM. The purpose of this research is to create a hybrid deep learning model that combines deep neural network-based models LSTM and Gated Recurrent Units (GRU) with 1D Convolutional Neural Networks (1D-CNN) that focuses on three stock prices, TSLA, GOOG, and TWTR. The findings reveal that the proposed model outperforms the LSTM network and other hybrid models in terms of prediction errors. This shows that the proposed model is effective in forecasting stock prices.

Keywords
Deep Learning, Time series, Forecasting, Stock prices, CNN, LSTM, GRU.

References
[1] Rob J. Hyndman, and George Athanasopoulos, Forecasting: Principles and Practice, OTexts, 2018.
[Google Scholar] [Publisher Link]
[2] Sima Siami-Namini, Neda Tavakoli, and Akbar Siami Namin, “A Comparison of ARIMA and LSTM in Forecasting time Series,” 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA), 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Li Deng, and Dong Yu, “Deep Learning: Methods and Applications,” Foundations and Trends in Signal Processing, vol. 7, no. 3–4, pp. 197–387, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Deepali M. Kotambkar, and Pallavi M. Wankhede, “Hybrid LSTM/GRU-based Domain Adaptation Model for Correlation Analysis to Detect Glaucoma,” SSRG International Journal of Electrical and Electronics Engineering, vol. 10, no. 1, pp. 168-175, 2023.
[CrossRef] [Publisher Link]
[5] Peter T. Yamak, Li Yujian, and Pius K. Gadosey, “A Comparison between Arima, Lstm, and Gru for Time Series Forecasting,” Proceedings of the 2019 2nd International Conference on Algorithms, Computing and Artificial Intelligence, pp. 49–55, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Mohil Maheshkumar Patel et al., “A Deep Learning-based Cryptocurrency Price Prediction Scheme for Financial Institutions,” Journal of Information Security and Applications, vol. 55, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Viswapriya Misra, “Time Series Forecasting with Applications to Finance,” NJIT Digital Common, 2021.
[Google Scholar] [Publisher Link]
[8] G. P. Dimf, P. Kumar, and K. Paul Joshua, “CNN with BI-LSTM Electricity Theft Detection based on Modified Cheetah Optimization Algorithm in Deep Learning,” SSRG International Journal of Electrical and Electronics Engineering, vol. 10, no. 2, pp. 35-43, 2023.
[CrossRef] [Publisher Link]
[9] Ioannis E. Livieris et al., “An Advanced CNN-LSTM Model for Cryptocurrency Forecasting,” Electronics, vol. 10, no. 3, p. 287, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Saul Alonso-Monsalve et al., “Convolution on Neural Networks for High-frequency Trend Prediction of Cryptocurrency Exchange Rates using Technical Indicators,” Expert Systems with Applications, vol. 149, p. 113250, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Kyung G.O, and I.F.Myung-Suck, “Evaluation of Short-Term Interval Models for Financial Time Series Forecasting,” SSRG International Journal of Industrial Engineering, vol. 2, no. 3, pp. 5-8, 2015.
[CrossRef]  [Publisher Link]
[12] Wenjie Lu et al., “A CNN-LSTM-based Model to Forecast Stock Prices,” Complexity, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Andres Vidal, and Werner Kristjanpoller, “Gold Volatility Prediction using a CNN-LSTM Approach,” Expert Systems with Applications, vol. 157, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Urooj Kaimkhani, Bushra Naz, and Sanam Narejo, “Rainfall Prediction Using Time Series Nonlinear Autoregressive Neural Network,” SSRG International Journal of Computer Science and Engineering, vol. 8, no. 1, pp. 30-38, 2021.
[CrossRef]  [Publisher Link]
[15] Hamed Habibi Aghdam, and Elnaz Jahani Heravi, Guide to Convolutional Neural Networks, Springer Cham, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Sumit Saha, A Comprehensive Guide to Convolutional Neural Networks — the ELI5 way, 2018. [Online]. Available: https://towardsdatascience.com/a-comprehensive-guide-to-convolutional-neural-networks-the-eli5-way-3bd2b1164a53