Modeling of Energy Consumption Based on Population: The Case of Turkey

In this study, Turkey's primary energy consumption (PEC) is modeled by linear regression analysis (LRA) based on population (P). The derived models are validated through some statistical approaches such as the determination coefficient, t-test, F-test, and residual analysis. For the performance assessment of the proposed model, three performance measures such as mean absolute percentage error (MAPE), root mean square error (RMSE) and mean absolute error (MAE) are used. Three scenarios are used for forecasting Turkey's PEC in the years 2013-2020. For each scenario, various assumptions are made considering different growth rate for P. Using the proposed model, Turkey's PEC is forecasted under different scenarios. The results show that the proposed model can be affectively used for forecasting of Turkey's PEC. The scenarios show that Turkey's PEC would vary between 135.523 Mtoe and 140.745 Mtoe in 2020.