A1150
Title: A time-varying causality approach for Botswana's trade balance and its determinants
Authors: Mpho Bosupeng - Griffith University (Australia) [presenting]
Abstract: The aim is to evaluate the causality between the trade balance and its determinants using the time-varying Granger causality approach proposed. A novel time-varying causality that is based on change detection algorithms is applied. Secondly, the new causality approach allows temporal fragilities in causal relationships to be examined through intensive subsample data analysis. The causality technique uses three algorithms to detect causality which are the forward recursive causality, rolling causality and the recursive evolving causality. Botswana depends on trading partners for the diamond trade and the time-varying causality approach will provide insights into how this reliance varies over the years. The results will assist policymakers in restructuring the economy, promoting domestic products, and measuring the effectiveness of macroeconomic policies. The time-varying Granger causality test is used to reveal the impact of domestic income, foreign income, and exchange rates on the trade balance. Botswana depends significantly on South Africa for imports and exports of diamonds to developed countries. It is expected this information will help policymakers in restructuring the economy and reaching high production capacities.
