StockGAN: Robust Stock Price Prediction Using GAN Algorithm.#
2. Affiliation:#
Department of International Business, National Taiwan University, Taipei, Taiwan.
3. Keywords:#
Stock price prediction, GAN algorithm, Deep learning.
4. Urls:#
Paper, Github: None.
5. Summary:#
(1): The research background of this paper is stock price prediction using deep learning techniques, particularly GAN algorithm.
(2): In the past, various methods have been proposed for stock price prediction; however, they lack accuracy and robustness. The authors are motivated to develop a robust model for stock price prediction using GAN algorithm.
(3): The research methodology proposed in this paper involves collecting stock data, preprocessing, feature extraction, and model training using GAN algorithm. The authors utilized features such as open, high, low, close, and volume to develop their model.
(4): The proposed model achieved good accuracy and a low error rate, shown with the metric score r2 with real predictions = 0.811166 and synthetic predictions = 0.674971. The MAE function produces real predictions = 0.020665, and synthetic predictions = 0.042406. The MRLE gains real = 0.001087 and synthetic = 0.002479. The results suggest that the GAN algorithm is a promising approach in dealing with accurate and dynamic stock prices.
6. Conclusion:#
(1): The significance of this work is to investigate the potential of using the GAN algorithm for robust stock price predictions. The authors aim to address the issue of accuracy and robustness of the traditional methods by proposing a new approach that can improve the accuracy of the predictions.
(2): Innovation point: The authors proposed a novel approach to address the issue of accuracy and robustness in stock price predictions using GAN algorithm. Performance: The results of the study demonstrate that the GAN algorithm is a promising approach to improve the accuracy of stock price predictions. The model achieved good accuracy and a low error rate compared with other techniques. Workload: The study has not provided any discussion on the workload of this approach, which is a potential limitation of the study.