Comparative Study of Conventional and Bayesian Methods for Determination of Factors Affecting Economic Growth in OIC

According to the growth literature, dozens of factors affect the growth and its process. The multiplicity of determinants of growth has caused uncertainty in the modeling and impeded consensus among the researchers. In the growth econometric theory, the theory and the chosen region dictate the effect of several specific factors. This paper has studied conventional approaches and Bayesian methods and compares these the way in which they work efficiently in determining the most effective determinants of economic growth in the Islamic cooperative countries. Therefore, Bayesian Model Averaging, Bayesian Maximum Likelihood, Fixed effects, Random effects and GMM in panel data have been used and the results have been compared. Due to differences in the economic structure of these countries, with other countries, in some area like industrialization and diversity of exports and social features, the capabilities of Bayesian approaches and conventional methods, were compared in the identification of growth factors and in this order we used data related to the Organization of Islamic cooperation countries during the period 1975-2015. According to the results, in a Bayesian space, and in the BMA approach, factors like saving rates, the credit of private sector, foreign direct investment, and exchange rate have put the robust impact on economic growth in these countries. In applying the BMA, investment has been ineffective, and the rest of the factors have affected economic growth in a different order. The results from conventional approaches are somewhat different from the results of Bayesian methods. The results of this research can be used in modeling economic growth and better management of the growth process in Islamic countries.