Bayesian Econometrics Approach in Determining of Effecting Factors on Pollution in Developing Countries (based on Environmental Performance Index)

Emphasis on sustainable development and the need to protect the environment as well as the adverse effects of environmental pollution on the quality of life have made environmental protection one of the main concerns of economic policymakers. For this purpose, approaches to improve the quality of the environment and the factors affecting it have triggered extensive theoretical and empirical studies over the past few decades. These issues have caught the attention of economic analysts. Accordingly, the main objective of this study is to investigate pollution determinants in developing countries from 1996 to 2016, using  Bayesian Model Averaging Method. Given the fact that the weighted mean square coefficient of GDP is positive, the Environmental Kuznets Curve (EKC) Hypothesis can be confirmed with a high degree of certainty. The probability of  this variable's effect is 0.98%, being partially a component of each of the 10 optimal models which highlights the great importance of this variable to explain the environmental performance. Energy consumption variables for each unit of GDP and value added of industry sector are placed in the second and third ranks with effectiveness  probability of 0.89 and 0.85, respectively. They also have a negative impact on environmental performance. Thus, energy consumption per unit of GDP is considered one of the elements of 8 out of 10 optimal models, while the value added of the industrial sector is an element of 7 out of 10 models. This highlights the relative importance of these variables in explaining environmental performance.