PERAMALAN HARGA EMAS MENGGUNAKAN METODE AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA)

Abstract

Gold has long been recognized as one of the most attractive investment instruments due to its function as a hedge against inflation and economic uncertainty. The volatility of gold prices, influenced by both global and domestic economic factors, necessitates accurate forecasting methods to support sound investment decision-making. This study aims to model and forecast gold prices in Indonesia using the Autoregressive Integrated Moving Average (ARIMA) method. The dataset consists of 156 weekly gold price observations covering the period from January 2022 to December 2024. The analysis was conducted through data exploration, stationarity testing, model order identification using ACF and PACF plots, and model selection based on the Akaike Information Criterion (AIC). The results indicate that the ARIMA(1,1,1) model provides the best fit with the lowest AIC value, and was therefore selected to generate forecasts for the subsequent 12 weeks. The forecasting results reveal a consistent upward trend in gold prices until mid-2025, although the prediction intervals widen over time, reflecting increasing uncertainty in long-term projections. These findings suggest that the ARIMA model is effective in capturing historical price patterns and can provide reliable insights into future price movements. The implications of this research highlight the model’s potential to guide investors and policymakers in developing short-term investment strategies, while future studies are recommended to incorporate external variables through multivariate approaches such as VAR or ARIMAX for more comprehensive forecasts.