- Position: PhD Student - Department of Economics
- Load: PhD Student Full Time
- Principal supervisor: Professor Geoffrey Kingston
- Associate supervisor: Professor Stefan Trueck
- Date of submission: 01/01/2003
- Thesis title: Real Estate Finance
- Purpose: The purpose of this paper is to examine optimal hedging strategies for currency-commodity portfolios by a model that incorporates regime switching dynamics into the correlation structure between gold, oil and exchange rates between the US dollar and commodity currencies such as the Australian dollar and the Canadian dollar.
- Originality: This is the first study to investigate bi-regime correlations and hedging applications for commodities and commodity currencies.
- Design/methodology/approach: We propose a new, extended version of Dynamic Conditional Correlation model of Engle (2002) that allows for regime shifts and the incorporation of exogenous variables in the evolution of conditional correlations.
- Findings: We find that the proposed model provides a good fit to the observed correlation structure between these assets. We apply the model to create a minimum variance hedge portfolio consisting of investments in oil, gold and commodity currencies. We test our approach against a number of benchmark models by comparing the sample volatilities of portfolio returns and find that the proposed model provides the best results with respect to the considered criteria.
- Research limitations/implications: Possible extensions in future work include the relaxation of the assumption of a multivariate normal distribution for modelling the asset returns as well as alternative models for the dependence structure such as copulas.
- Keywords: Dynamic Hedging, Optimal Hedging ratio, DCCX model, Markov regime switching, State space model