Effects of multicollinearity on electricity consumption forecasting using partial least squares regression

Abstract

The electricity forecasting is important for accurate investment planning of energy production/generation; however energy is also essential input for economical and industrial development. In this study, we aim to discuss the impact of economic growth on annual electricity consumption. It is true that economic variables usually have correlations with each with variable degrees. Partial least squares regression (PLSR) is one of the effective ways of dealing with the multicollinearity problem especially arises in several econometric models. In situations when there are a large number of highly correlated explanatory variables, decomposition by PLSR can be used to select a small number of linear combinations of the original variables which explain the great amount of covariance between explanatory and response variables. Thus, the aim of this study is to perform the partial least squares regression method to forecast annual electricity consumption using historical data for Turkey. (C) 2012 Published by Elsevier Ltd. Selection and/or peer review under responsibility of Prof. Dr. Huseyin Arasli