SEARCH WITHIN CONTENT
Citation Information : Statistics in Transition New Series. Volume 17, Issue 3, Pages 557-574, DOI: https://doi.org/10.21307/stattrans-2016-038
License : (CC BY 4.0)
Published Online: 06-July-2017
The aim of the paper is to search for hedges and safe havens within three instrument classes: assets (represented by the S&P500 index), gold and oil prices, and dollar exchange rates. Weekly series of returns of all the instruments from the period January 1995 – June 2015 are analysed. The study is based on conditional correlations between the instruments in different market regimes obtained with the use of copula-DCC GARCH models. It is assumed that different market regimes will be identified by statistical clustering techniques; however, only conditional variances (without conditional covariances) will be taken into account. The reason for this assumption is connected with the fact that variances can be understood as market risk, and, as such, are a good indicator of market conditions. A considerable advantage of such an approach is the lack of need to determine the number of market regimes, as it is established by clustering quality measures. What is more, the methodology used in the paper makes it possible to treat the relations between instruments symmetrically. The results obtained in the study reveal that only dollar exchange rates can be treated as a (strong) hedge and a (strong) safe haven for other instruments, while gold and oil are a hedge for assets.