ARIMA with Regression Model in Modelling Electricity Load Demand

Abstract

Electricity is among the most crucial needs for every people in this world. It is defined by the set of physical phenomena related with the flow of electrical charge. The importance of electricity itself leads to the increasing electricity load demand in the world including Malaysia. The purpose of the current study is to evaluate the performance of combined ARIMA with Regression model in forecasting electricity load demand in Johor Bahru. Box-Jenkins Autoregressive Integrated Moving Average (ARIMA) and Regression models will be used as benchmark models since the model has been proven in many forecasting context. Using Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) as a forecasting accuracy criteria, the study concludes that the combined method is more appropriate model.