IBM Stock Forecast Using LSTM, GRU, Attention and Transformer Models

Document Type

Conference Proceeding

Publication Date



In the stock market, the change of stock price has always been the most concerned thing of shareholders. However, due to the uncertainty of the stock market, it is also very difficult to predict the trend of stock price. In this paper, first, we collected IBM's stock price data from January 2, 1962 to August 21, 2017, and then we calculated the middle price of the stock. Next, we present four model approach to stock prediction. The method used in our study is known as LSTM, LSTM+GRU, Attention, and Transformer. We carried out several sets of experiments to test the validity of the analysis. The test results show that the LSTM and GRU model superposition method is more effective than other methods in predicting the stock price trend of IBM.

Publication Title

2023 IEEE International Conference on Control, Electronics and Computer Technology, ICCECT 2023

First Page Number


Last Page Number




This document is currently not available here.