Abstract:

In this study, energy efficiency determinants of Turkey were investigated using TOPSIS (Technique for Order Similarity to Ideal Solution) method, Tobit model and Artificial Neural Network (ANN) algorithm. The study performed a two stage analysis: calculation of TOPSIS energy efficiency scores and estimation with artificial neural networks and Tobit models. In the first stage, Annual energy efficiency scores of Turkey for the period 1960-2013 calculated by using  carbon emissions, Gross Domestic Product (GDP), renewable and non-renewable energy consumption, labor force and capital stock variables. In the second stage, the efficiency scores obtained by the TOPSIS method used as a dependent variable in YSA and Tobit models, while carbon emissions, GDP, renewable and non-renewable energy consumption, labor force, capital stock/labor ratio, dummy variables representing crisis years, deterministic trend and square of trend used as independent variables. The results exhibit that the ANN model can predict the experimental results with high correlation coefficient, 0.998.

Keywords: