The Application of the Multivariate GARCH Models on the BRICS Exchange Rates

Abstract

The study investigated the BRICS exchange rate volatility using the Multivariate GARCH models. The study used the monthly time series data for the period January 2008 to January 2018. The BEKK-GARCH model revealed that all the variables were found to be statistically significant. The diagonal parameters estimates showed that only Russia and South Africa were statistically significant. This implied that the conditional variance of Russia and South Africa’s exchange rates are affected by their own past conditional volatility and other BRICS exchange rates past conditional volatility. The BEKK-GARCH model also revealed that there is a bidirectional volatility transmission between Russia and South Africa. The results from the DCC-GARCH model revealed that Brazil, China, Russia and South Africa had the highest volatility persistence and India has the least volatility persistence. All the BRICS exchange rates show that the fitted residuals are not normally distributed except for Russia. The recommendations for future studies were articulated.

Multivariate Garch Models Comparison in Terms of the Symmetric and Asymmetric Models

“…There are multiple equations and variables in it. Various exogenous and endogenous variables may also be present, according to Metsileng et al (2020). Although it might not be as flexible as linear models, the Multivariate GARCH model is intended to represent some robust patterns in the data set.…”

“…Ijomah and Enewari (2020) looked at the volatility transmission between the price of oil and the exchange rate using multivariate GARCH modelling. BRICS (Brazil, Russia, India, China, and South Africa) currency rate volatility was examined by Metsileng et al (2020) using Multivariate GARCH modelsFrancq and Zokoian (2015) examined various Multivariate volatility models' estimation methods. Bala and Takimoto (2017) investigated stock return volatility spill-overs in emerging and developed markets using the Multivariate GARCH (M-GARCH) model and its variations.…”