Energy demand prediction is very important to timely supply, market regulation, exports targeting and energy security. Different methods introduced to energy demand prediction, due to volatilities and non-linearity on energy demand, non-linear techniques have good results among them. Neural networks and genetic algorithm are well-known and most widely used techniques in this field that both of them have own strength and weaknesses. Imposing the specific form, necessity to the large samples and weakness on global optimum finding are important weaknesses on each method which these shortcomings can be fixed by combining them. In this study real coded genetic algorithm is used for neural network training as a hybrid algorithm (RGA- NN). After applying and comparing this technique with common techniques on energy demand prediction between 1967 -2011, the results confirm higher predictive performance of hybrid technique and the explanatory power of the used variables.
Sadeghi, H., Sohrabi Vafa, H., & Nouri, F. (2013). Applications of Neural Network Based on Genetic Algorithm for Long Term Energy Demand Forecasting. Quarterly Journal of Applied Theories of Economics, 1(2), 29-52.
MLA
Hossein Sadeghi; Hossein Sohrabi Vafa; Fatemeh Nouri. "Applications of Neural Network Based on Genetic Algorithm for Long Term Energy Demand Forecasting". Quarterly Journal of Applied Theories of Economics, 1, 2, 2013, 29-52.
HARVARD
Sadeghi, H., Sohrabi Vafa, H., Nouri, F. (2013). 'Applications of Neural Network Based on Genetic Algorithm for Long Term Energy Demand Forecasting', Quarterly Journal of Applied Theories of Economics, 1(2), pp. 29-52.
VANCOUVER
Sadeghi, H., Sohrabi Vafa, H., Nouri, F. Applications of Neural Network Based on Genetic Algorithm for Long Term Energy Demand Forecasting. Quarterly Journal of Applied Theories of Economics, 2013; 1(2): 29-52.
)