Research on Stock Price Prediction Based on LSTM Hybrid Model

DOI: https://doi.org/10.62517/jse.202511609

Author(s)

Tongshu Wu

Affiliation(s)

Dept. of AI and Big Data, Xi'an Jiaotong-Liverpool University. Suzhou, China

Abstract

Stock price forecasting is an important topic in financial time series analysis. Traditional time series models have limitations in processing nonlinear financial data, and deep learning models show application potential due to their strong feature learning capabilities. In this study, a long short-term memory network and an autoregressive comprehensive moving average model are constructed to systematically compare their performance differences in stock price prediction. The study adopts a unified technical index system and a multi-dimensional evaluation framework to evaluate the performance of the model from three dimensions: prediction accuracy, direction judgment and generalization ability. The empirical results show that the deep learning model has significant advantages in capturing the nonlinear characteristics of stock prices, and is better than the traditional time series method in all evaluation indicators. It is manifested in higher prediction accuracy, better trend judgment ability, and stronger out-of-sample generalization performance. The research results confirm the effectiveness of deep learning in financial time series analysis and provide a methodological reference for quantitative investment strategies. The comparative analysis framework established by this study can be extended to other financial prediction scenarios and has high practical application value.

Keywords

Component; Stock Price Prediction; Long Short-Term Memory Network; Time Series Analysis; Deep Learning; ARIMA Model; Machine Learning

References

[1] W. Lu, J. Li, Y. Li, A. Sun, and J. Wang, "A CNN-LSTM-based model to forecast stock prices," Complexity, vol. 2020, article ID 6622927, 2020. [2] A. Moghar and M. Hamiche, "Stock market prediction using LSTM recurrent neural network," Procedia Computer Science, vol. 170, pp. 1168-1173, 2020. [3] A. Moghar and M. Hamiche, "Stock market prediction using LSTM recurrent neural network," Procedia Computer Science, vol. 170, pp. 1168-1173, 2020. [4] Gao, Y., Wang, R., & Zhou, E. (2021). Stock Prediction Based on Optimized LSTM and GRU Models. Scientific Programming, 2021, Article ID 4055281. https://doi.org/10.1155/2021/4055281 [5] S. M. Al-Selwi, M. F. Hassan, S. J. Abdulkadir, A. Muneer, E. H. Sumiea, A. Alqushaibi, and M. G. Ragab, "RNN-LSTM: From applications to modeling techniques and beyond—Systematic review," Journal of King Saud University-Computer and Information Sciences, vol. 36, p. 102068, 2024. [6] X. Wen and W. Li, "Time Series Prediction Based on LSTM-Attention-LSTM Model," IEEE Access, vol. 11, pp. 48322-48331, 2023. [7] N. You, L. Han, D. Zhu, and W. Song, "Research on Image Denoising in Edge Detection Based on Wavelet Transform," Appl. Sci., vol. 13, no. 3, p. 1837, Jan. 2023. [8] X. Yu, Z. Chen, Y. Ling, S. Dong, Z. Liu, and Y. Lu, "Temporal Data Meets LLM-Explainable Financial Time Series Forecasting," arXiv preprint arXiv:2306.11025, 2023.