<![CDATA[Inflation is a key economic indicator that affects purchasing power, economic growth, and financial stability. Accurate forecasting is essential for policymakers to implement effective monetary and fiscal policies. However, traditional models like ARIMA (Autoregressive Integrated Moving Average) mainly capture general trends and often fail to address inflation volatility. This study enhances inflation forecasting accuracy by applying the ARIMA-GARCH hybrid model, which combines trend estimation with volatility modelling. Focusing on Indonesia’s inflation patterns using recent data, it addresses a gap in existing research. This type of research uses quantitative methods, and the data were obtained from the official website of Bank Indonesia. The dataset consists of 240 monthly Indonesian inflation data points spanning from September 2004 to August 2024. The ARIMA (0,1,1)-GARCH (2,0) model is used to analyze inflation trends and volatility dynamics. The model evaluation shows strong predictive performance, with a Mean Absolute Percentage Error (MAPE) of 2.73% and Root Mean Squared Error (RMSE) of 0.74 for training data. Testing data results in a MAPE of 18.95% and RMSE of 0.702, which remains within an acceptable range. These findings highlight the importance of incorporating volatility modelling in inflation forecasting to enhance economic decision-making. A reliable forecast mitigates economic uncertainty, thereby providing a stronger foundation for achieving long-term economic growth. This study contributes by demonstrating the practical application of ARIMA-GARCH in Indonesia’s inflation modelling, providing valuable insights for policymakers in managing inflation-related risks.]]>
