Issue |
ITM Web Conf.
Volume 70, 2025
2024 2nd International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2024)
|
|
---|---|---|
Article Number | 02023 | |
Number of page(s) | 7 | |
Section | Machine Learning in Healthcare and Finance | |
DOI | https://doi.org/10.1051/itmconf/20257002023 | |
Published online | 23 January 2025 |
Forecasting Stock Prices with Artificial Intelligence
Course of Professional Study (CPS), Northeastern University, Boston, MA 02115, U.S.A.
Corresponding author: zhao.danx@northeastern.edu
The purpose of this study is to investigate the closing prices of stocks in Artificial intelligence. The objective is to enhance the accuracy of future stock price Prediction to support investment or trading decisions. The models used in this paper include Simple Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), LSTM with peephole connectivity, and Gated Recurrent Unit (GRU). To conduct the study, Wal-Mart stock data is utilized to accurately predict future stock prices. The results show that the MSE for the SimpleRNN test is higher, indicating weaker generalization. The MSE of the basic LSTM test is lower than that of the RNN, indicating stronger generalization. The validated and tested MSEs of LSTM with peephole connectivity are higher than the basic LSTM and GRU. GRU performs as well as the basic LSTM but has the lowest LSTM training MSE. This stock prediction task requires the GRU model, which is the most suitable choice based on the training time.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.