Forecasting Stock Market Prices Through Real-Time Stock Data Utilizing Deep Learning Techniques

Accurate financial forecasting is challenging due to the unpredictable nature of the stock market. Research in stock rankings prediction and portfolio development has gained significance in today’s globalized financial landscape. When predicting stock prices using stock technical indicators (STIs) and financial data, dimensionality reduction is a crucial initial step. This study integrates dimensionality reduction and deep learning techniques, applying Principal Component Analysis (PCA) for dimensionality reduction and Convolutional Neural Network (CNN) for prediction. The research evaluates six STIs using historical stock data from HDFC Bank and ICICI Bank sourced from Yahoo Finance. Assessment metrics include mean square error (MSE), mean absolute error (MAE), root mean square error (RMSE), and coefficients of determination (R2) for regression analysis. Results show the PCA-CNN model outperforms alternative models (GA-ANN, PCA-ENN, and CNN) across all parameters, highlighting the importance of integrating dimensionality reduction and deep learning for accurate stock price predictions. In the dynamic global financial markets, adopting advanced models is essential for informed investment decisions and optimized portfolio management.