Bivariate Error Correction FIGARCH and FIAPARCH Models on the Australian All Ordinaries Index and Its SPI Futures.
journal contributionposted on 05.06.2017 by Dark, Jonathan
Any type of content formally published in an academic journal, usually following a peer-review process.
In this paper we extend the univariate FIGARCH and FIAPARCH models to a bivariate framework. We estimate bivariate error correction FIGARCH and FIAPARCH models between the All Ordinaries Index and its SPI futures using constant correlation and diagonal parameterisations. We therefore employ a flexible estimation approach that captures the long run equilibrium relationship between the two markets, bi-directional return causality, long memory and asymmetries in volatility, and time varying correlations. The results strongly support the use of this approach. Strong bi-directional return causality exists with the index bearing the burden of adjustment to deviations from long run equilibrium. The results also illustrate the importance of allowing for long memory, asymmetries in volatility, and time varying correlations.