Metoda Autoregressive untuk Peramalan Jangka Panjang

Abstract

Changes in seasonal patterns in Indonesia are closely related to rainfall. Various forecasting techniques were developed to produce better accuracy. In this study ARIMA linear forecasting techniques were used. The data used is secondary data from BMKG Kalianget Station, Sumenep from January 2008 - December 2017 with a monthly rainfall research variable. To measure the accuracy of the forecast results used by RMSE.From the result of this study, ARIMA ([1,6],0,0)(0,1,1)¹²providing better accuracy than ARIMA (1,0,0)(0,1,1)¹² for predicting the next 1 month or 12 months (a year ahead).