<![CDATA[The rapid development of information technology and data mining has encouraged the use of machine learning algorithms in various fields, including the financial sector and capital markets. One of the main challenges in stock price prediction is the large number of available variables, not all relevant to the target variable, potentially reducing accuracy and causing overfitting. This study aims to analyze the benefits of applying feature selection in improving the performance of the Random Forest Regression algorithm for stock price prediction. The dataset used in this research consists of ten years of historical stock price data from PT Aneka Tambang Tbk (ANTM). The research was conducted using an experimental approach by developing two models: (1) Random Forest Regression without feature selection and (2) Random Forest Regression with feature selection using the Spearman Correlation method. Model performance was evaluated using Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Square Error (RMSE), Coefficient of Determination (R²), and Mean Absolute Percentage Error (MAPE). The experimental results show that the model with feature selection achieved better performance, with improvements in all evaluation metrics, such as reduced error values (MAE: 26.22; RMSE: 51.82; MAPE: 1.32%) and increased R² (0.9895). These findings confirm that integrating feature selection with Random Forest Regression can improve prediction accuracy, reduce model complexity, and minimize overfitting risk. Therefore, feature selection plays a significant role in enhancing the effectiveness of machine learning models in stock price prediction.]]>
