STOCK PRICE PREDICTION: A COMPARATIVE STUDY USING LINEAR REGRESSION, RANDOM FOREST, AND LSTM MODELS

This study proposes a comparative comparison of three distinct prediction models—Linear Regression, Random Forest, and Long Short-Term Memory (LSTM)—for projecting stock prices of 29 firms, including the S&P 500 index, from January 1, 2000, to June 27, 2024. The study seeks to assess the efficacy and precision of these models by the analysis of historical stock data and the computation of critical metrics, including Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and R-squared (R²). The results show that although the Random Forest model surpasses Linear Regression, the LSTM model exhibits enhanced prediction performance owing to its capacity to capture temporal relationships in time-series data. This study enhances the domain of financial forecasting by demonstrating the efficacy of several machine learning models in stock price prediction and proposing paths for further research.