Research on Stock Price Prediction based on LSTM Modeling-A Stock Market as a Case Study

Yuxin Wang

doi:10.54097/8gq1ja53

Authors

Yuxin Wang

DOI:

https://doi.org/10.54097/8gq1ja53

Keywords:

LSTM; SSE 50 Index; stock price prediction; time series analysis

Abstract

This research intends to predict the closing values of the Shanghai Stock Exchange 50 Index using an enhanced Long Short-Term Memory (LSTM) neural network approach. The investigation leverages a comprehensive data set from August 2013 to August 2023, encompassing various market conditions. Alongside the primary LSTM analysis, the study introduces supplementary methodologies, specifically the Auto-correlation Function (ACF) and Partial Auto-correlation Function (PACF), to validate and enrich the time-series analysis. This study verified the non-stationary of the original data and processed them with method. 80%of the data was allocated for training, while the subsequent 20% was reserved for testing, the LSTM model, with its intricate architecture, exhibits superior predictive capabilities, particularly in stable market conditions. However, its performance wanes during volatile market periods, suggesting further refinements are necessary. This research contributes to the existing literature by applying machine learning methodologies to the turbulent and complex A-share market, offering invaluable insights for both academic researchers and market investors.

Downloads

Download data is not yet available.

References

Agrawal J G, Chourasia D V S, Mittra D A K. State-of-threat in stock prediction techniques. International Journal of Advanced Research in Electrical Electronics & Instrumentation Engineering, 2013, 2(4).

Kao L J, Chiu C C, Lu C J, Yang J L. Integration of nonlinear independent component analysis and support vector regression for stock price forecasting. Neurocomputing, 2013, 534-542.

Alkhatib K. et al. A New Stock Price Forecasting Method Using Active Deep Learning Approach. Journal of Open Innovation: Technology, Market, and Complexity, 2022, 8(2): 2199-8531.

Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput, 1997, 9(8).

Liu S, Liao G and Ding Y. Stock transaction prediction modeling and analysis. 2018 13th IEEE Conference on Industrial Electronics and Applications (ICIEA), 2018, 2787-2790.

Naik N, Mohan B R. Study of Stock Return Predictions Using Recurrent Neural Networks with LSTM, International Conference on Engineering Applications of Neural Networks. 2019.

Wu. et al. A graph-based CNN-LSTM stock price prediction algorithm with leading indicators. Multimedia Systems, 2021, 29(3).

Mehtab S and Sen J. Stock Price Prediction Using CNN and LSTM-Based Deep Learning Models. Proceedings of the 4th IEEE International Conference on Electronics, Communication, and Aerospace Technology, 2020, 1481-1486.

Millikan E, Subramanian Pand Joseph M H. Bitcoin Vision: Using Machine Learning and Data Mining to Predict the Short-Term and Long-Term Price of Bitcoin. Webology, 2021, 18: 751-760.

Prasad A and Seetharaman A. Importance of Machine Learning in Making Investment Decision in Stock Market. Vikalpa, 2021, 46(4): 209–222.