Peramalan dengan Metode Seasonal Autoregressive Integrated Moving Average (SARIMA) di Bidang Ekonomi

(Studi Kasus: Inflasi Indonesia)

Abstract

A present event is probably a reiteration from a past event. The reiteration of an event every particular time period indicates seasonal pattern. Seasonal Autoregressive Integrated Moving Average (SARIMA) is one of the methods that is used for data forecasting which has seasonal pattern. The purposes of this research are finding out the best SARIMA model and forecasting the inflation in Indonesia for period January 2016 until December 2016 using the best SARIMA model. Sample of this research is 96 Indonesia inflation data (mtm) for period January 2008 until December 2015. The technique of this research is purposive sampling. There are five steps of SARIMA method, those are model identification, model estimating, diagnostic checking, selecting the best model, and forecasting. Based on the analysis, the best SARIMA model is SARIMA (1,0,0)(0,1,0)¹². The forecasting of Indonesia inflation 2016 has similar pattern with the previous time. The inflation increases in January 2016 and decreases in February 2016 until April 2016. The inflation increases again in Mey 2016 until August 2016 and decreases in September 2016 until November 2016. At last, the inflation increases in December 2016.