An Effective Integrative CNN-GRU Framework for Stock Price Forecasting

In the modern period, due to technological advancements, most people are attracted to the stock market to generate profits. Due to a lack of adequate knowledge about the stock market, they face losses. Predicting stock prices is a difficult task due to the complex nature of the market and the factors that can impact it. Some factors that influence the stock market are technical and psychological. Deep learning models have recently demonstrated promising results in capturing complex patterns and interrelationships within financial time series data. This abstract introduces a novel approach to predicting stock prices by combining convolutional neural networks (CNNs) and gated recurring units (GRUs). The suggested architecture combines deep learning models to take advantage of each model’s distinct capabilities in stock price prediction. The GRU component detects temporal correlations and long-term patterns, whereas the CNN component concentrates on extracting spatial information from historical data. Future stock price predictions are more precise thanks to the combination of CNN and GRU, which enables a thorough representation of the data.