<![CDATA[Stock investment is very attractive to investors because it provides many benefits. The higher the profit offered in investing, the higher the risk investors will face. Stock price analysis is needed to predict stock prices to reduce the risk of loss. Stock data has dynamic, non-linear, and unpredictable characteristics, so a method is required to overcome the limitations of time series data. The Support Vector Regression method will be applied in this study to predict PT. Adaro stock prices. The Support Vector Regression method is one method that does not require assumptions, so it can be used to overcome the limitations of regression analysis with time series data. The problem often faced when using the SVR method is determining the optimal hyperparameters. This study determines the optimal hyperparameter using the grid search algorithm. The data is divided into training and testing data with a ratio of 90:10. The best kernel to predict the share price of PT. Adaro is using a linear kernel with a value of Cost = 4 and epsilon = 0.0001. The model produces MAPE testing data of 1.608%, which means the model is very good for prediction.]]>
Copyrights © 2025
