Encoder Decoder Generative Adversarial Network Model for Stock Market Prediction
This is an incremental improvement for financial analysts and traders, addressing specific bottlenecks like mode collapse and unstable training in GANs for stock prediction.
The paper tackles stock price forecasting by proposing a GRU-based Encoder-Decoder GAN (EDGAN) model, which achieves superior forecasting accuracy and training stability compared to traditional GAN variants in volatile markets.
Forecasting stock prices remains challenging due to the volatile and non-linear nature of financial markets. Despite the promise of deep learning, issues such as mode collapse, unstable training, and difficulty in capturing temporal and feature level correlations have limited the applications of GANs in this domain. We propose a GRU-based Encoder-Decoder GAN (EDGAN) model that strikes a balance between expressive power and simplicity. The model introduces key innovations such as a temporal decoder with residual connections for precise reconstruction, conditioning on static and dynamic covariates for contextual learning, and a windowing mechanism to capture temporal dynamics. Here, the generator uses a dense encoder-decoder framework with residual GRU blocks. Extensive experiments on diverse stock datasets demonstrate that EDGAN achieves superior forecasting accuracy and training stability, even in volatile markets. It consistently outperforms traditional GAN variants in forecasting accuracy and convergence stability under market conditions.