FORECASTING INDONESIA’S ELECTRICITY CONSUMPTION PER CAPITA USING ARIMA MODEL

Precise electricity consumption predictions play an important role in managing electricity and in developing the economy. Forecasting electricity usage therefore is imperative. This study examined the per capita electricity consumption of Indonesia in order to find the appropriate ARIMA model for the variable estimate. The data series used electricity consumption per capita in Indonesia from 1980 to 2014.In the assignment of appropriate (p, d, q) values in ARIMA, the parameters for ACF and PACF were taken into account as well as stationary data. The optimal model is then determined by selecting the one with the lowest Akaike Information Criterion (AIC), the Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the Mean Absolute Percentage Error (MAPE). The simulation findings showed that the ARIMA (1,1,0) model was statistically suitable to predict electricity consumption per capita in Indonesia.