Prediction of electrical energy consumption in Cameroon through econometric models

Abstract

Electricity consumption is on the rise in developing countries. Most energy demand forecasting research studies aim to provide sufficient electricity forecast for accurate planning of investments in electricity generation and distribution. The main objective of this study is to develop efficient and realistic solutions for forecasting electricity consumption in Cameroon. This article proposes a hybrid model based on vector error correction models (VECM) and Holt-Winters exponential smoothening (HES), combines from an improved algorithm of the gradient descend in order to minimize the forecast error. The prediction results of this hybrid approach are compared to similar studies in the literature, to artificial intelligence models as well as to predictions from official published scientific work. Also, it is applied to forecast the future net electricity consumption of Cameroon until 2024. The results of the study indicate that the proposed model can generate more realistic and reliable forecasts. It can also be argued that it responds better to some unexpected reactions from the time series.