In this article, we construct a portfolio of commodity futures which mimics the Dow Jones Commodity Index and perform an extensive stress testing exercise with a focus on hybrid scenarios. The increased volume of investments in commodities as financial instruments over the last decades underline the importance of a more thorough framework for stress testing of related portfolios. Our study is the first to show comparatively the marginal impact of the model choice for portfolio components versus the marginal role of tail dependence on the portfolio profit and loss in stress testing exercises. We model the distribution of returns of portfolio components with an asymmetric AR-GARCH model combined with Extreme Value Theory for extreme tails, and employ multivariate copula functions to model the time-varying joint dependence structure. Our study reveals that indeed, for a realistic stress test, a special attention should be given to the tail risk in individual commodity returns as well as to tail correlations. We also draw conclusions about parameter risk persistent in stress testing exercises. Finally yet importantly, in line with Basel IIIb, the study highlights the importance of using forward-looking hybrid and hypothetical scenarios over historical scenarios. Keywords Stress testing • Commodity futures • Risk measures • Extreme value theory • Copula functions Research funded by Adolf Øiens Donasjonsfond, "Energizing new Computional Frontiers" and by the
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.