Multivariate Analysis of the Volatility Transmission of Commodity Market and Financial Markets in Nigeria
L. E. George *
Ignatius Ajuru University of Education, Rumuolumeni, Port Harcourt, Rivers State Nigeria.
I. D. Essi
Rivers State University, Port Harcourt, Rivers State, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
This study examined the multivariate volatility transmission among selected commodity and financial markets in Nigeria, using daily observations on Brent crude oil, copper, gold, natural gas, exchange rate and stock market returns covering December 29, 2009 to November 7, 2025. The study examined the volatility and covariance transmission structure, estimated own-market and cross-market volatility spillovers, and assessed volatility persistence across the selected markets. The study adopted an ex-post facto research design and employed VECH-MGARCH and BEKK-GARCH models to capture the dynamic behaviour of conditional variances, covariances and volatility transmission. The results revealed substantial volatility clustering and non-normality in all the return series, with natural gas recording the highest standard deviation, while gold recorded the lowest. The VECH-MGARCH results showed significant own-market volatility effects across the markets, with the Nigerian stock market exhibiting the strongest immediate response to recent shocks, while the GARCH components indicated considerable persistence in volatility. Cross-market effects were also observed, including positive and negative transmission patterns among selected commodity, exchange-rate and stock-market pairs. The BEKK-GARCH results further confirmed significant own-market ARCH effects, ranging from 0.2159 for gold to 0.3037 for the stock market, while GARCH effects ranged from 0.8788 for natural gas to 0.9323 for gold. The combined persistence estimates exceeded unity across all six markets, ranging from 1.0986 for natural gas to 1.2235 for the exchange rate, indicating strong persistence but also raising an important model-stability consideration. Despite the evidence of persistent volatility and cross-market covariance dynamics, the estimated conditional correlations were generally very small, with the largest reported value being 0.0366 between copper and gold, suggesting weak contemporaneous linear co-movement among the markets. The study therefore concludes that the selected Nigerian commodity and financial markets exhibit substantial volatility persistence and identifiable cross-market transmission.
Keywords: Multivariate GARCH, volatility transmission, volatility persistence, brent crude oil, copper, gold, natural gas, exchange rate, Nigerian stock market, conditional correlations