Optimized Stock Price Forecasting: A Deep Learning Approach Using QCTO Based GRU Network

Stock represents ownership in a publicly traded company and stock price is the current market value of a company’s stock. It fluctuates based on factors such as supply and demand, company performance, and overall market conditions. Predicting accurate stock price is the basic task of every trader and investor to gain profit from the financial market. Traditional trend prediction models are generally based on experts knowledge and experience. Often, it is difficult to find out hidden patterns contained in the stock time series data due to its inherent volatility and complexity. To address this issue, a Quantum Class Topper Optimization based Gated Recurrent Unit (QCTO-GRU) model is proposed to predict stock price movement, highlighting its ability to capture intricate patterns in dynamic market conditions. The parameters of GRU model are optimized using QCTO, a meta-heuristic optimization technique. The proposed model excels in predicting the stock price and outperforms the existing models with the value of 0.01271 and 0.00975 for MAE and MSE respectively.