The prediction of stock prices is a critical area of research in finance due to its significant implications for investors, financial institutions, and the broader economy. In order to predict stock values, this study thoroughly evaluates a number of machine learning approaches, including Decision Tree, Random Forest, Lasso, Ridge, Elastic Net, Support Vector Machine (SVM), and Linear Regression. Evaluating these models’ performance in terms of computational efficiency and prediction accuracy is the main goal. Our empirical analysis reveals that while each model exhibits unique strengths and weaknesses, ensemble techniques, particularly those that integrate Linear Regression and Elastic Net, demonstrate superior performance. The proposed ensemble model leverages the individual merits of Linear Regression and Elastic Net, achieving enhanced accuracy and robustness in capturing and forecasting complex patterns in stock price data. The study’s findings help investors make better judgments by offering insightful information about how to choose the best models for financial prediction tasks.