Modelling Real Exchange Rate Volatility in a Developing Country

Abstract

Existing studies have provided empirical evidence on volatility clustering on exchange rate. This presents a situation of uncertainty and risks for future outcomes. This paper investigated the presence and nature of real exchange rate volatility in the Ghanaian Economy. This study would inform and guide policy-makers on currency risks and currency crises management. A Breusch-Pagan test for ARCH effects was performed. Further, an ARCH(1) and GARCH(1,1) processes were explicitly modelled to measure volatility. The major empirical and methodological contribution of this study is the explicit modelling of the conditional mean and conditional variance processes. The GARCH(1,1) model was the right model for exchange rate risk modelling in Ghana. The exchange rate regime change from fixed to floating caused a spike in volatility from 1983 to 1986. Over the period, there have been intermittent spikes in volatility indicating that Ghana’s international competitiveness deteriorated over the period of study. However, from 2001 to 2010, the volatility has been minimal.

Keywords: GARCH, Uncertainty, Volatility clustering, Exchange rate, Ghana.