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

Fu, Sihan; Tang, Zefan; Li, Jialin

doi:10.1109/iccect57938.2023.10140896

Cited by 4 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then the input gate controls the proportion of input that is allowed to enter the memory cell. In the final step, the output gate determines the value that gets produced [11].…”

Section: Long Short-term Memory (Lstm)mentioning

confidence: 99%

Stock prediction and analysis based on machine learning algorithms

2024

ACE

View full text Add to dashboard Cite

The stock market has consistently remained a focal point of substantial concern for investors. Nevertheless, due to the intricate, tumultuous, and often noisy nature of the stock market, forecasting stock trends presents a formidable obstacle. To augment the accuracy of stock trend predictions, the author adopts a combination of the Long Short-Term Memory (LSTM) neural network and a noise reduction technique known as Ensemble Empirical Mode Decomposition (EEMD). This composite model is employed to develop predictions for the daily stock price increases, aiming to provide more precise insights into market behavior. The framework is capable of generating the daily stock price change trend curve based on the training outcomes. EEMD, standardization, and other data preprocessing methods can effectively reduce the noise of the stock market. In this paper, three U.S. stocks from 2010 to 2023 are chosen as the research subjects. After the training is completed, the prediction curve generated by the model closely aligns with the actual curve. Furthermore, three commonly used evaluation metrics were utilized to assess the models performance. Based on all those experimental outcomes, this model adeptly forecasts the stocks trend.

show abstract

“…Then the input gate controls the proportion of input that is allowed to enter the memory cell. In the final step, the output gate determines the value that gets produced [11].…”

Section: Long Short-term Memory (Lstm)mentioning

confidence: 99%

Stock prediction and analysis based on machine learning algorithms

2024

ACE

View full text Add to dashboard Cite

show abstract

“…Unlike RNN-based models, the Transformer leverages a multi-head self-Attention mechanism to grasp global positional relationships, thereby strengthening its ability to learn longterm dependencies. Fu et al conducted several experiments to evaluate the efficacy of the Transformer model in predicting IBM's stock price trend [11]. Ding et al introduced a new methodology utilizing the Transformer with several enhancements [1].…”

Section: Literature Reviewmentioning

confidence: 99%

“…Common methods include taking the dot product of their vectors, calculating the cosine similarity of their vectors, or employing additional neural networks for evaluation. In this study, we adopt the approach of first calculating the dot product of the two vectors, followed by applying the tanh activation function to obtain the similarity: (11) In the second stage, a calculation method Softmax normalization is introduced to transform the scores obtained in the first stage. On one hand, it allows normalization by organizing the original score calculations into a probability distribution where the sum of all element weights is equal to 1.…”

Section: Attentionmentioning

confidence: 99%

Unveiling the Predictive Power: Comparative Analysis of Cutting-Edge Deep Learning Models for Stock Price Forecasting

Guan,

Zhu,

Xiao

2023

HBEM

View full text Add to dashboard Cite

Accurate stock price prediction plays a fundamental role in informing government financial regulations and facilitating effective arbitrage strategies for investors. With the application of deep learning algorithms in finance, significant progress has been made to improve the accuracy of stock price prediction. In this paper, first, we collected stock price data from four listed companies from different sectors. Then, we used four competitive methods for prediction, namely LSTM, GRU-LSTM, Attention-LSTM and Transformer-LSTM. The validity of the study is supported by multiple sets of comparative experiments. Our experimental results show that LSTM shows superiority in predicting stock prices, while Transformer-LSTM model has better generalization ability.

show abstract

Enhancing Stock Price Predictions Through LSTM-based Recurrent Neural Networks

Krishna Kumar,

Natarajan,

et al. 2024

2024 International Conference on Integrated Circuits and Communication Systems (ICICACS)

View full text Add to dashboard Cite

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

Cited by 4 publications

References 15 publications

Stock prediction and analysis based on machine learning algorithms

Stock prediction and analysis based on machine learning algorithms

Unveiling the Predictive Power: Comparative Analysis of Cutting-Edge Deep Learning Models for Stock Price Forecasting

Enhancing Stock Price Predictions Through LSTM-based Recurrent Neural Networks

Contact Info

Product

Resources

About