Dynamic conditional correlation models and their application to portfolio risk mitigation

This bachelor thesis investigates asymmetry in returns of corn, gold and crude oil (both spot and futures) and hedging effectiveness of these commodities when employing DCC family models for hedge ratio estimation. The asymmetry in conditional variances was found to be significant only in case of crude oil spot and futures returns and asymmetry in conditional correlation of spot and futures returns was not shown to be significant in neither of the investigated commodities. With respect to the hedging performance, we conclude that differences in hedging performance measured by hedging effectiveness index are negligible and thus do not support superiority of DCC family models over OLS, which served as a benchmark. Historical Value at Risk, on the contrary, identified the DCC with asymmetry in conditional variance (despite asymmetry not being significant) to be appropriate for corn hedging, however not for the other two commodities, where the OLS based hedge ratio performed similarly or even better than the DCC family models. The main contribution of the thesis thus lays in empirical investigation of asymmetry in returns of selected commodities and testing hedging potential of DCC family based hedge ratio.