ESTIMASI VOLATILITAS TINGKAT INFLASI BULANAN INDONESIA MENGGUNAKAN MODEL GARCH(1,1)

Abstract

. Inflation is one of the key macroeconomic indicators used to assess a country's economic stability, making the analysis of its volatility essential in supporting monetary and fiscal policy formulation. This study aims to estimate the volatility of Indonesia’s monthly inflation using the GARCH(1,1) model without conducting any forecasting stage. The dataset consists of monthly inflation rates from January 2006 to June 2025, obtained from the Central Bureau of Statistics (BPS). Stationarity was tested using the Augmented Dickey-Fuller (ADF) test, while model parameters were estimated using the Maximum Likelihood Estimation (MLE) method implemented in Python. The results show that the GARCH(1,1) model effectively captures the dynamics of inflation volatility, particularly during economic shock periods such as 2008 and 2013. The high value of α₁ indicates that short-term shock effects are the main driver of inflation volatility, while the low β₁ value suggests limited long-term persistence. However, the Ljung-Box diagnostic test reveals remaining autocorrelation in the residuals, implying that the GARCH(1,1) model has not fully accommodated all volatility patterns. Therefore, this model can serve as a baseline for volatility analysis and may be extended using more advanced models such as EGARCH or GARCH-M to achieve higher accuracy. These findings provide empirical insights into Indonesia’s inflation volatility dynamics and serve as a reference for more responsive price stabilization policies.