Comparative Analysis of Linear Regression, Random Forest Regressor and LSTM for Stock Price Prediction

Stock price prediction is an important topic of research in financial markets and many machine learning algorithms are used to estimate future stock prices. This study compares three prominent stock price prediction algorithms: Linear Regression, Random Forest, and Long Short-Term Memory (LSTM). The main objective of the research is to assess these algorithms' predictive ability using R-squared (R2), Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE) metrics. The study uses historical data from ten S&P 500 and ten NIFTY 50 stocks to represent a wide range of industries and market circumstances. This data is used to train and test each algorithm, and the evaluation criteria indicated above are used to determine each method’s performance. The study’s findings offer an insight into the ability of Linear Regression, Random Forest, and LSTM models in predicting stock values across markets and sectors.