Analysis of the Generalized Autoregressive Conditional Heteroskedasticity (GARCH) Model in Forecasting Antam Gold Prices in Indonesia

This research is a type of applied research with a quantitative approach to analyze the results of Antam's gold price forecasting in Indonesia with the best Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model obtained. GARCH is a model used to analyze data volatility over time, especially in forecasting data that frequently experiences fluctuations. The research data used is 234 weekly historical data on Antam gold prices in Indonesia in the period January 2020 to June 2024. The results of this research show that by using the best ARIMA (2,1,2) GARCH (1,2) model, gold price forecasting results are obtained. Antam in Indonesia for the period July 2024 to June 2025, namely IDR 1,376,096 to IDR 1,781,239. This model has a high level of accuracy in forecasting Antam's gold price in Indonesia with a Mean Absolute Percentage Error (MAPE) value of 1.13%.