A frequentist and Bayesian regression analysis to daily peak electricity load forecasting in South Africa

A frequentist and Bayesian regression analysis to a piecewise linear regression model for daily peak electricity load forecasting in South Africa for the period 2000 to 2009 is discussed in this paper. The developed model captures a wide variety of electricity demand drivers such as temperature, seasonal, lagged demand and calendar effects. A Bayesian analysis provides a way of taking into account uncertainty in the estimation of the piecewise linear regression parameters. Uncertainty about the true values of the Bayesian parameter estimates is incorporated into the analysis through the use of a non-informative prior distribution. The results obtained are easy to explain to management. Empirical results showed that an increase in electricity peak demand, if temperature decreases by 1°C, could be any value between 140 and 200 MW during the winter months. Similarly during the summer months the increase in electricity peak demand, if temperature increases by 1°C, ranges from -20 to 80 MW. There is a persistent increase of around 2 MW in hourly electricity peak demand with time in South Africa. Electricity demand in South Africa is more sensitive to the winter period. Demand for electricity during holidays decreases significantly compared to a day before and after a holiday. This information and the quantification of such uncertainty are important for load forecasters in the power Utility Company in South Africa (Eskom) as it helps them in the determination of consistent and reliable supply schedules.   Key words: Posterior distribution, temperature, piecewise linear regression, load forecasting.