Hybrid Deep Learning Model for Stock Price Prediction: Evidence from Guizhou Moutai Stock

Jingyi Lu

2024

Abstract

Research on stock price prediction has been increasingly important, particularly as financial markets get more complicated. This study evaluates the execution of various hybrid deep learning models for stock price prediction using data from Guizhou Moutai, a leading Chinese company, from January 5, 2015, to September 6, 2024. The models compared include MLP, RNN, CNN, LSTM, BiLSTM, CNN-LSTM, CNN-BiLSTM and CNN-BiLSTM-AM. In contrast to alternative approaches, the CNN BiLSTM-AM method is better suited for stock price prediction and offers investors a dependable means of making stock investing selections. CNN, BiLSTM, and AM make up this technique. It achieves the lowest RMSE (21.273) and MAPE (0.944%) while maintaining the highest R² value (0.9580), highlighting its superior predictive accuracy. This study provides a more reliable tool for data-driven decision-making in the financial market, which not only helps improve the accuracy of investors' decisions, but also encourages the use of deep learning in the finance industry. Furthermore, this study offers a valuable reference for further investigations into maximizing model performance, cutting down on computational expenses, and integrating external macroeconomic variables.

Download


Paper Citation


in Harvard Style

Lu J. (2024). Hybrid Deep Learning Model for Stock Price Prediction: Evidence from Guizhou Moutai Stock. In Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI; ISBN 978-989-758-726-9, SciTePress, pages 243-250. DOI: 10.5220/0013214200004568


in Bibtex Style

@conference{ecai24,
author={Jingyi Lu},
title={Hybrid Deep Learning Model for Stock Price Prediction: Evidence from Guizhou Moutai Stock},
booktitle={Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI},
year={2024},
pages={243-250},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013214200004568},
isbn={978-989-758-726-9},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI
TI - Hybrid Deep Learning Model for Stock Price Prediction: Evidence from Guizhou Moutai Stock
SN - 978-989-758-726-9
AU - Lu J.
PY - 2024
SP - 243
EP - 250
DO - 10.5220/0013214200004568
PB - SciTePress