Forecasting electricity consumption: A comparison of models for New Zealand
Forecasting electricity consumption is of national interest to any country. Future electricity forecasts are not only required for short and long term power planning activities but also in the structure of the national economy. This paper proposes six forecasting models developed for electricity consumption in New Zealand. Three of these models (Logistic, Harvey Logistic and Harvey) are based on growth curves. A further model uses economic and demographic variables in multiple linear regression to forecast electricity consumption while another uses these factors to estimate future saturation values of the New Zealand electricity consumptions and combine the results with a growth curve model. The sixth model makes use of the Box-Jenkins ARIMA modeling technique. The developed models are compared using goodness of fit, forecasting accuracy and future consumption values. The future consumptions are also compared with the available national forecasts. The comparisons revealed that the best overall forecasts are given by the Harvey model for both the Domestic and the Total electricity consumption of New Zealand while a specific form of the Harvey model, the Harvey Logistic model, is the best in forecasting Non- Domestic electricity consumption.