TITLE: Advanced Stock Market Prediction Using Hybrid GRU-LSTM Techniques

Computer Science Engineering

The stock market is influenced by various factors like economic conditions, investor behavior, and global events, which cause frequent fluctuations in stock prices. This makes accurate predictions difficult because the market is highly unpredictable. As a result, forecasting stock prices becomes a challenging task. Existing models like ARIMA, LSTM, and GRU are widely used for stock price prediction. ARIMA is effective for linear data. LSTM and GRU are better at handling complex non-linear data that makes them more effective for stock market predictions. But these models are complex and time-consuming to implement and require significant computational resources. To address the challenges in existing methods, a new technique called Regularized GRU-LSTM is introduced. This method combines the strengths of both LSTM and GRU to improve performance. LSTM is a type of neural network that is used to remember important information over time, which makes it suitable for sequential data, and GRU is simple and faster while handling sequential data effectively. This model not only improves prediction accuracy but also reduces time complexity in processing stock time series data. This approach demonstrates superior performance compared to stand-alone GRU, LSTM, and ARIMA models, facilitating efficient and accurate short-term stock price forecasting and advancing the field of financial time series analysis.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.