Machine Learning Based Modelling of Economic Growth and Quality of Governance: The MENA Region

Abstract Governance (which entails transparency, accountability, the rule of law and the presence of effective and legitimate institutions) is considered an essential factor in economic development. A large number of academic studies have attempted to identify and explain the influence of governance quality on economic growth, bearing on different theoretical perspective and using a panoply of estimation methods, including correlation and regression analyses. This study approaches the phenomenon from a predictive analytical perspective using contemporary Machine Learning techniques to uncover the most important predictors of economic growth in the MENA region in sample observed from 1996 to 2020. Random Forest algorithm was used with three machine learning models (Support Vector Machine, Boosted TREE, Linear Regression) to predict economic growth. The empirical results indicated that the predictions obtained using Random Forest were more accurate than those obtained by the other models. The results indicated that Government Effectiveness, Control of Corruption and Rule of Law are the most influential factors explaining economic growth.