Is Africa’s current growth reducing inequality? Evidence from some selected african countries

The main objective of this study is to model and predict the real GDP rate using Bayesian approach. A Bayesian VAR (BVAR), a Bayesian linear model and switching regime Bayesian models were employed for the real GDP rate, inflation rate and interest rate. From the set of variables that were connected to real GDP, for identifying the most relevant ones using the data for Romanian economy, we applied the selection algorithm based on stochastic search. Weight of revenues in GDP, weight of budgetary deficit in GDP, investment rate and inflation rate are the most correlated variables with the real GDP rate. The averages of posterior coefficients of models were used to make forecasts. For Romania on the horizon 2011-2014, the unrestricted switching regime models generated the most accurate forecasts.