Portfolio Stress Testing Applied to Commodity Futures

Abstract

In this thesis, 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. Limitations to the risk management practices became clear during the recent financial crisis, and the increased volume of investments in commodity markets 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 the marginal impact of the model choice for portfolio components versus the marginal role of tail dependency in stress testing exercises to assess the portfolio risk profile. We model the distribution of portfolio components with an asymmetric GARCH model combined with Extreme Value Theory for extreme tails. We then implement a copula function to model the time-varying joint dependency structure. Our study reveals that indeed, for an accurate stress test, a special attention should be given to the tail risk in individual commodity returns as well as to tail correlations. Furthermore, we find that the parameter risk in the model for the individual portfolio components impact the portfolio profit and loss profile most in stress testing exercises. Finally yet importantly, in line with Basel III, the study highlights the importance of using forward-looking hybrid and hypothetical scenarios over historical scenarios, for a comprehensive stress testing