STOCK PRICE PREDICTION USING MACHINE LEARNING

Abstract

Accurate stock price prediction plays a vital role in financial decision-making by enabling investors to identify market trends, optimize investment strategies, and minimize potential risks. The increasing availability of historical financial data has encouraged the adoption of machine learning techniques for analyzing complex stock market behavior. However, conventional statistical forecasting approaches often struggle to capture nonlinear relationships, high market volatility, and dynamic price fluctuations, resulting in limited predictive performance. Historical stock market data obtained from publicly available financial sources covering multiple years, including features such as opening price, closing price, high price, low price, trading volume, and adjusted closing price, are utilized for model development. The collected data undergo preprocessing involving data cleaning, handling missing values, feature selection, normalization, and train-test splitting to improve model reliability and learning efficiency. Multiple machine learning models, including Linear Regression, Decision Tree Regression, Random Forest Regression, Support Vector Regression, and XGBoost Regression, are implemented and comparatively analyzed to estimate future stock prices. Model performance is evaluated using Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and the coefficient of determination (R² Score). Among the evaluated approaches, the XGBoost Regression model achieves the highest prediction accuracy with the lowest error values and superior R² performance. The developed system demonstrates improved forecasting capability and provides an efficient framework for intelligent stock market trend analysis and price prediction.