Multivariate dependence risk and portfolio optimization: An application to mining stock portfolios

This study proposes an integrated framework to model and estimate relatively large dependence matrices using pair vine copulas and minimum risk optimal portfolios with respect to five risk measures within the context of the global financial crisis. We apply this methodology to two 20-asset mining (gold and iron ore-nickel) sector portfolios from the Australian Securities Exchange. The pair vine copulas prove to be powerful tools for the modeling of changing dependence risk under three different period scenarios combined with the optimization of portfolios that have complex patterns of dependence. The portfolio optimization results converge, on average, in some stocks.