Modeling conditional correlations for risk diversification in crude oil markets

Chia-Lin Chang, Michael McAleer, Roengchai Tansuchat

Abstract

ABSTRACT

This paper estimates univariate and multivariate conditional volatility and conditional correlation models of spot, forward and futures returns from three major benchmarks of the international crude oil markets, namely Brent,West Texas Intermediate and Dubai, to aid with the process of risk diversification. Conditional correlations are estimated using Bollerslev's constant conditional correlation model, Ling and McAleer's vector autoregressive moving average-generalized autoregressive conditional heteroscedasticity (VARMA-GARCH) model, the vector autoregressive moving average-asymmetric generalized autoregressive conditional heteroscedasticity (VARMA-AGARCH) model of McAleer et al and a dynamic conditional correlation model by Engle. The paper also presents the autoregressive conditional heteroscedasticity and generalized autoregressive conditional heteroscedasticity effects for returns and shows the presence of significant interdependencies in the conditional volatilities across returns for each market.