Double Seasonal ARIMA Model for Forecasting Load Demand

Abstract

This study investigates the use of a double seasonal ARIMA model for forecasting load demand. For the purpose of this study, a one-year half hourly Malaysia load demand from 1 September 2005 to 31 August 2006 measured in Megawatt (MW) is used. The mean absolute percentage error (MAPE) is used as the measure of forecasting accuracy. We use Statistical Analysis System, SAS package to analyze the data. Using the least squares method to estimate the coefficients in a double SARIMA model, followed by model validation and model selection criteria, we propose ARIMA(0; 1; 1)(0; 1; 1)48(0; 1; 1)336 with in-sample MAPE of 0.9906% as the best model for this study. Comparing the forecasting performances by using k-step ahead forecasts and one-step ahead forecasts, we found that the MAPE for the one-step ahead out-sample forecasts from any horizon ranging from one week lead time to one month lead time are all less than 1%. We thus propose that a double seasonal ARIMA model with one-step ahead forecast as the most appropriate model for forecasting the two-seasonal cycles Malaysia load demand time series. Keywords: Load forecasting; double seasonal ARIMA model; k-step ahead forecast; one-step ahead forecasts.