Inflation Forecasting by Commodity Using the Autoreggressive Integrated Moving Average (ARIMA) Method

Abstract

The aim of this research is to determine the best Autoregressive Integrated Moving Average (ARIMA) model and its implementation to predict monthly inflation in Indonesia. The data used is the inflation data on the expenditure group of foods, beverages, cigarettes and tobaccos in period January 2010 until December 2019. The method used in this study is the documentation technique. The data analysis technique used is the Autoregressive Integrated Moving Average (ARIMA) which is calculated using the SPSS version 26. The result of this research shows that ARIMA model (12,0,12) is the best model to predict monthly inflation on the expenditure group of foods, beverages, cigarettes and tobaccos in Indonesia for the next period. The results of forecasting 12 months in 2020 with the ARIMA model (12,0,12), in January until April decrease, then for May until August increase while September decrease and in October until December experienced an increase. Therefore, inflation is considered a major problem in the modern economy so that inflationary forecasting can be used in making an economic policy of the coming period which aims to reduce and stabilize price growth. Keywords—forecasting, inflation, Autoreggressive Integrated Moving Average (ARIMA)