Modelling Exchange Rate Volatility Of KES/USD Using Garch Family Models

Abstract

Investors, Policy makers, Governments etc. are all consumers of exchange rates data and thus exchange rate volatility is of great interest to them. In this paper, the importance of exchange rate volatility modelling has been brought out in the problem statement and the significance of the study. The paper thus seeks to study the KES/USD exchange rate volatility using GARCH family models. Symmetric and asymmetric models have been used to capture volatility characteristics of exchange rates on data spanning 23/10/1993 to 21/03/2014 and a comparison is made under four different conditional distribution assumptions i. e. Normal, Skewed Normal, Student-t and Skewed Student-t distributions. Investigations conducted showed that the asymmetric EGARCH model with skewed student distribution emerges as the best based on AIC and log-likelihood. It can account for the asymmetry in the exchange rate return series while the skewed student-t accounts for some of the skewness and leptokurtosis. The data is further split into four sub-periods and volatility characteristics analyzed per period. Each period shows a significant occurrence of high fluctuations in the exchange rate.