Application of Asymmetric-GARCH Type Models to The Kenyan Exchange Rates
Loading...
Date
2023-08-31
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
EJ-MATH, European Journal of Mathematics and Statistics
Abstract
Modelling and forecasting the volatility of a financial time series has become
essential in many economic and financial applications like portfolio optimization and risk
management. The symmetric-GARCH type models can capture volatility and leptokurtosis.
However, the models fail to capture leverage effects, volatility clustering, and the thick tail
property of high-frequency financial time series. The main objective of this study was to apply
the asymmetric-GARCH type models to Kenyan exchange to overcome the shortcomings of
symmetric-GARCH type models. The study compared the asymmetric Conditional
Heteroskedasticity class of models: EGARCH, TGARCH, APARCH, GJR-GARCH, and
IGARCH. Secondary data on the exchange rate from January 1993 to June 2021 were obtained
from the Central Bank of Kenya website. The best fit model is determined based on parsimony
of the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Log Likelihood criterion, and minimisation of prediction production errors (Mean error [ME] and
Root Mean Absolute error [RMAE]). The optimal variance equation for the exchange rates data
was APARCH (1,1) - ARMA (3,0) model with a skewed normal distribution (AIC = -4.6871, BIC
= -4.5860). Volatility clustering was present in exchange rate data with evidence of the leverage
effect. Estimated Kenya’s exchange rate volatility narrows over time, indicating sustained
exchange rate stability