Comparative Analysis of Stock Price Prediction Using Machine Learning

The exchange of stocks, bonds, and other assets on the stock market is a complicated and dynamic process that is impacted by several variables. Machine learning methods have drawn more and more interest lately in financial forecasting. With the unavoidable uncertainty and volatility in the prices of stocks, investors keep seeking techniques to foresee future trends to avert losses and maximize probable profits. The purpose of this comparison study is to assess how well four machine learning models—Long Short-Term Memory (LSTM), Random Forest, Support Vector Regression (SVR), and Linear Regression—predict Oracle stock prices. We evaluated the model using metrics like RMSE, MSE, and MAE using a dataset with 9,648 samples collected between December 3, 1986, and June 21, 2024. According to the findings, SVR is the best-performing model with the lowest error rates, making it the preferred choice for stock price prediction. LSTM also performed well, closely matching SVR. While Linear Regression produced reasonably accurate results, the Random Forest model did not perform as effectively and showed notably poor performance, suggesting that it may not be suitable for this particular dataset or problem.