Penerapan Model Seasonal Autoregressive Fractionally Integrated Moving Average Pada Data Inflasi di Indonesia

Abstract

Current inflation data is influenced by previous inflation data. Inflation data from time to time is indicated to have a long memory and seasonal pattern. The Seasonal Autoregressive Fractional Integrated Moving Average (SARFIMA) model is one of the models used to predict data that has a long memory and seasonal pattern. The purpose of this research was to find out the the best SARFIMA model and forecast inflation in 2018 using the best SARFIMA model. The sample in this research was Indonesian monthly inflation data for the period January 2008 to December 2017. There are four stages of SARFIMA modeling, namely model identification, parameter estimation, diagnostic checking, and application of models for forecasting. Based on the results of the analysis, the best SARFIMA model selected based on the AIC and MSE criteria is the SARFIMA model with d = 0.687. The results of inflation forecasting from January to December 2018 show a fluctuating value every month with the inflation rate at 3.30% - 3.65%.