Modeling sector wise demand for electricity in Sri Lanka using bayesian techniques

Abstract

The demand for the electricity in Sri Lanka depends mainly on the activities of domestic, industrial, and commercial sectors. Electricity has become the most important source of energy in the domestic sector of the country that is desirous of achieving the newly developed nation status. This study aims to identify a model to forecast Electricity demand. The unpredictable changes in the activities cause the demand of electricity to change unusually. This makes predicting the demand very difficult sometimes. Therefore in this research the way of incorporating changes in the activities in the time series model for forecasting is used, which is known as Bayesian Forecasting (BF) model. The Comparison of this model with the Autoregressive Integrated Moving averages (ARIMA) model, which is generally used, was made. According to the calculated mean absolute percentages error (MAPE), BF model gave better performance in term of higher degrees accuracy of forecasting. Thus on one hand it is encouraging that the Sri Lanka electricity authorities can have some faith in the model used for forecasting. Hence the chosen econometric work does have a considerable impact of the policy decisions in the Sri Lankan electricity supply industry.