Author(s)

YunChen Jiang

Affiliation(s)

College of Intelligent Engineering, Xi'an Jiaotong-Liverpool University, Suzhou 215000, Jiangsu, China.

Corresponding Author

YunChen Jiang

ABSTRACT

We think stock price is a long-term research topic and find that stock price forecasts are very important. And found that the changes in the stock price often show no linear and the current traditional stock price forecast can not well reflect the forecast results. So the deep learning method to predict stock prices. And explored three of the mainstream methods, RNN, working principles of the CNN and LSTM models. We then tried to use three models for TSLA, NFLX, LLY, AAPL share price data from four companies in different sectors. Finally, by comparing the MSE obtained by different companies through different models. The MAE and MAPE standards were compared. Finally, the most applicable general model is obtained. The experiment and final comparison found that the RNN model is higher than the other two models. At the end of the study, we considered the RNN model as a general model with higher comprehensive ability to predict stock prices in different fields.

KEYWORDS

CNN; RNN; LSTM; Forecast stock price

CITE THIS PAPER

YunChen Jiang. Comparison of the CNN, RNN and LSTM models for high-frequency stock price forecasts. Journal of Trends in Financial and Economics. 2024, 1(1): 7-12. DOI: 10.61784/jtfe3004.

REFERENCES

[1] Ding G, Qin L. Study on the prediction of stock price based on the associated network model of LSTM. International Journal of Machine Learning and Cybernetics, 2020, 11(6), 1307-1317.

[2] Jahan I, Sajal S. Stock price prediction using recurrent neural network (RNN) algorithm on time-series data. In 2018 Midwest instruction and computing symposium, 2018.

[3] Selvin S, Vinayakumar R, Gopalakrishnan EA, et al. Stock price prediction using LSTM, RNN and CNN-sliding window model. In 2017 international conference on advances in computing, communications and informatics (icacci) , 2017, 1643-1647.

[4] Li Haitao. The Markov forecasting method is used to predict stock prices. Statistics and Decision-making, 2002, (5): 25-26.

[5] Prado HD, Ferneda E, Morais LCR, et al. On the Effectiveness of Candlestick Chart Analysis for the Brazilian Stock Market. Procedia Computer Science, 2013, 22: 1136-1145.

[6] Lin Y, Guo H, Hu J. An SVM-Based Approach for Stock Market Trend Prediction. The 2013 International Joint Conference on Neural Networks (IJCNN), 2014, 4-9.

[7] Mizuno H, Kosaka M, Yajima H, et al. Application of Neural Network to Technical Analysis of Stock Market Prediction. Studies in Informatic and Control, 1998, 7: 111-120

[8] Kumar M, Thenmozhi M. Forecasting Stock Index Movement: A Comparison of Support Vector Machines and Random Forest. Social Science Electronic Publishing, 2006.

[9] Zhang Rumeng, Zhang Huamei. Comparative analysis of BP neural network and ARMA-GARCH model in stock prediction. Journal of Science, 2021: 41(4): 14-20.

[10] Li Yong. Design and Implementation of an Intelligent Stock Prediction System Based on Deep Neural Networks: [Master's Thesis] Xi'an: Northwest University, 2019.

[11] Nikou M, Mansourfar G, Bagherzadeh J. Stock Price Prediction Using Deep Learning Algorithm and Its Comparison with Machine Learning Algorithms. Intelligent Systems in Accounting, Finance and Management, 2019, 26: 164-174. DOI : 10.1002/isaf.1459.

[12] Yang Qi, Cao Xianbing. Analysis and prediction of stock prices based on the ARMA-GARCH model. The Practice and Understanding of Mathematics, 2016, 46(6): 80-86.

[13] Huang Ying, Yang Huijie. Financial time series prediction based on the XGBoost and LSTM models. Technology and Industry, 2021, 21(8): 158-162.

[14] Zhang Kanglin. Analysis and prediction of the stock price by the LSTM model based on pytorch. Computer Technology and Development, 2021, 31(1): 161-167.

[15] Song Gang, Zhang Yunfeng, Bao Fangxun. Stock prediction model based on particle swarm optimization of LSTM. Journal of Beijing University of Astronautics, 2019, 45(12): 2533-2542.

[16] Chen W, Chai KY, Lau CT, et al. Leveraging Social Media News to Predict Stock Index Movement Using RNN-Boost. Data & Knowledge Engineering, 2018, 118: 14-24. DOI: 10.1016/j.datak.2018.08.003.

[17] Maqsood H, Mehmood I, Maqsood M, et al. A Local and Global Event Sentiment Based Efficient Stock Exchange Forecasting Using Deep Learning. International Journal of Information Management, 2020, 50: 432-451. DOI: 10.1016/j.ijinfomgt.2019.07.011.

[18] Song Gang, Zhang Yunfeng, Bao Fangxun, et al. A stock prediction model based on particle swarm optimization LSTM, 2019.

[19] Cui Guochao. Research on the Characteristics of Neural Network Models and Packet Size. Wireless Internet Technology, 2012 (3): 105.

[20] Gu Yongpeng, Qin Dibo, Zhang Xiang, et al. Tourist prediction based on LSTM Advances in Applied Mathematics, 2023, 12: 2143.

[21] Qiao Ruoyu. Neural network-based stock prediction model. Operations research and Management, 2019, 28 (10): 132-140.