Forecasting Hourly Electricity Demand in Egypt Using Double Seasonal Autoregressive Integrated Moving Average Model

摘要

Egypt has faced a major problem in balancing electricity produced and electricity consumed at any time in the day. Therefore, short-term forecasts are required for controlling and scheduling of electric power system. Electricity demand series has more than one seasonal pattern. Double seasonality of the electricity demand series in many countries have considered. Double seasonality pattern of Egyptian electricity demand has not been investigated before. For the first time, different double seasonal autoregressive integrated moving average (DSARIMA) models are estimated for forecasting Egyptian electricity demand using maximum likelihood method. DSARIMA (3, 0, 1) (1, 1, 1)_(24) (2, 1, 3)_(168) model is selected based on Schwartz Bayesian Criterion (SBC). In addition, empirical results indicated the accuracy of the forecasts produced by this model for different time horizon.