Modelling Dynamic Conditional Correlations in Spot, Forward and Futures Returns

Volatility (or risk) is a key variable in many areas of finance, and there are many applications that require an accurate estimate of volatility. One important application is in designing optimal dynamic hedging strategies. Engle (1982) proposed an autoregressive conditional heteroscedasticity (ARCH) model, which allows the conditional variance to change over time. This model has been extended over time, and has proved to be empirically successful in explaining the behaviour of stock returns, commodity prices and exchange rates. In my dissertation I will examine optimal hedging strategies for various commodities, exchange rates and stock indices. I will apply recently developed dynamic conditional correlation models as well as their static counterparts, in order to examine whether the dynamic models can be used to develop improved hedging strategies.