Analysis of the logistic model for predicting New Zealand electricity consumption

The ability of the Logistic model for forecasting electricity consumption is investigated with application to New Zealand data. The fitting of logistic curves to the historical data uses a Fibonacci search technique to establish optimum asymptotes. This paper analyses the effectiveness of the Logistic forecast with the historical predictions made by the Electricity Sector Forecasting Committee (ESFC) of the Ministry of Energy, New Zealand over the subsequent period of deregulation in the electricity industry. The comparison has revealed that in general the Logistic model forecasts have given rise to more accurate forecasts than those using economic data and that deregulation has had an effect on electricity use. New forecasts of the Logistic model, Ministry of Economic Development (Statistics Unit) and forecasts published jointly by Sinclair Knight Merz and CAE (Centre for Advanced Engineering, University of Canterbury, New Zealand) are compared for the years 2002- 2017