Gamma Regression Approach for Determining Factors of Inflation in Indonesia

Abstract

High inflation in Indonesia poses challenges to macroeconomic stability by reducing investment, increasing the risk of financial crises, and slowing economic growth. Although inflation has been widely studied, limited research has compared the suitability of conventional multiple regression and Generalized Linear Model (GLM) gamma regression for modeling inflation data with non-normal characteristics. This study aimed to examine the effects of the Bank Indonesia policy rate, exchange rate, and export value on Indonesia’s inflation rate while identifying the more appropriate analytical model. A quantitative associative research design was employed using secondary monthly data collected from the official publications of Statistics Indonesia and Bank Indonesia for the period January 2019 to December 2024, resulting in 72 observations. The data were analyzed using multiple linear regression and gamma regression based on the Generalized Linear Model framework, with model comparison performed to determine the most suitable approach. The findings showed that gamma regression provided a better fit than conventional multiple regression for explaining inflation. The gamma regression results indicated that the Bank Indonesia policy rate and export value significantly affected inflation, whereas the exchange rate had no significant effect. These findings demonstrate the importance of selecting an analytical model that matches the characteristics of the data to improve the accuracy of inflation analysis. The study contributes to the application of Generalized Linear Models in macroeconomic research and provides evidence that may support policymakers in developing more effective inflation control strategies.