PERFORMANCE COMPARISON OF RANDOM FOREST REGRESSION, SVR MODELS IN STOCK PRICE PREDICTION

The stock market is characterized by high volatility and complexity, making it an intriguing and challenging subject for researchers and practitioners. This study aims to predict stock prices by comparing the performance of two machine learning models: Random Forest Regression and Support Vector Regression (SVR). These models were selected for their ability to handle complex data and high volatility. The dataset used in this study consists of BNI stock data over the last five years (2019–2024), comprising a total of 1,211 data points. Testing was conducted using a cross-validation approach, and model performance was evaluated based on several metrics, including MSE, R², RMSE, MAPE, MAE, and Score. The results indicate that Random Forest Regression outperforms SVR. The model achieved an MAE of 17.766, an RMSE of 22.376, and an R² of 0.997. These findings suggest that Random Forest Regression is more effective in predicting stock prices, particularly in unstable market conditions. This study recommends Random Forest Regression as a reliable model for stock price prediction, with potential applications in other stock markets with similar characteristics.