A Time Varying Parameter State-Space Model for Analyzing Money Supply-Economic Growth Nexus

摘要

In this paper, we propose a time-varying parameter state spacemodel for analyzing predictive nexus of key economic indicators suchas money supply and Gross Domestic Product (GDP). Economic indicatorsare mainly used for measuring economic trends. Policy makers inboth advanced and developing nations make use of economic indicatorslike GDP to predict the direction of aggregate economic activities.We apply the Kalman filter and Markov chain Monte Carlo algorithm toperform posterior Bayesian inference on state parameters specified froma discount Dynamic Linear Model (DLM), which implicitly describesthe relationship between response of GDP and other economic indicatorsof an economy. In our initial exploratory analysis, we investigatethe predictive ability of money supply with respect to economic growth,using the economy of Nigeria as a case study with an additional evidencefrom South African economy. Further investigations reveal that leading variables like capital expenditure, the exchange rate, and thetreasury bill rate are also useful for forecasting the GDP of an economy.We demonstrate that by using these various regressors, there isa substantial improvement in economic forecasting when compared tounivariate random walk models.