PREDIKSI HARGA EMAS UNTUK PENGAMBILAN KEPUTUSAN INVESTASI MENGGUNAKAN ALGORITMA CART (CLASSIFICATION AND REGRESSION TREE)

Abstract

The dynamic fluctuations in gold prices, influenced by various global economic factors, make gold price prediction an important topic in finance and investment. This study aims to analyze and predict gold prices using the Classification and Regression Tree (CART) algorithm, which is a decision tree–based machine learning method. Historical gold price data from 2013 to 2023 were used as training data, while data from 2024 to 2025 were used for testing. The research process includes preprocessing, data splitting, CART model construction, and performance evaluation using metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and R-squared (R²). The results indicate that the CART model is capable of providing reasonably accurate and effective predictions of gold prices, making it a viable tool for investment decision-making. With its simplicity and interpretability, CART helps uncover patterns in gold price data and offers valuable estimates for investors and market participants.