Olushina Olawale Awe (Obafemi Awolowo University):

In this paper, we present a robust Bayesian technique for modeling Economic Indicators by developing and implementing a flexible Bayesian approach for dynamic economic variables, with application to the Nigerian Economy. Economic indicators are macroeconomic variables that indicate or reveal the direction of growth of an economy. They tell us how well an economy is performing or thriving. Increase in the GDP of a country, for instance, is generally perceived as an indicator of improvement in the standard of living of its inhabitants. Economic indicators are mainly used for measuring economic trends. Policy makers in both advanced and developing nations make use of economic indicators like GDP to predict the direction of aggregate economic activities. We present a new MCMC algorithm to obtain posterior inference on the state space parameters specified from a Third Order Bayesian Dynamic Model (BDM) which implicitly describes the variables that can be thought of as the overall state of an economy. Upon estimating this new model using economic data from 1960-2009, our method has the quality of being able to detect outliers, structural breaks and historical trend in the time series considered. Our initial exploratory analyses indicate that traditional leading variables like Money Supply, Exchange Rate and Capital Expenditure can prove useful in forecasting the long-run economic growth in Nigeria of which GDP serves as a proxy. Our method provides a full option that can facilitate posterior simulation through MCMC. Our Model is also capable and good for forecasting, detecting abrupt changes and handling outliers in high frequency economic time series.