Time varying and dynamic correlation among oil and natural gas prices: Multivariate long memory approach

This study examines the interdependence of natural gas price (HENRYHUB) and oil prices (WTI and BRENT). The aim of this paper is to examine how the dynamics of correlations between the markets evolved from January 01, 2004 to February 26, 2015. To this end, we adopt a dynamic conditional correlation (DCC) model into a multivariate Fractionally Integrated Asymmetric Power ARCH (FIAPARCH) framework, which accounts for long memory, power effects, leverage terms and time varying correlations. The empirical findings indicate the evidence of time-varying comovement, a high persistence of the conditional correlation and the dynamic correlations revolve around a constant level and the dynamic process appears to be mean reverting. Moreover, the univariate FIAPARCH models are particularly useful in forecasting market risk exposure for synthetic portfolios of stocks and currencies.