Expected Shortfall Modelling of the CARBS Indices

Abstract

The purpose of this study is to make use of time-varying volatility models to estimate expected shortfall (ES) for the CARBS indices and a global minimum variance portfolio (GMVP) constructed using the CARBS indices. The GARCH, GJR-GARCH and EGARCH models are considered. Furthermore, six different distributional assumptions are made regarding the error distribution. The evidence suggests that skewness and kurtosis are important factors to consider when modelling financial returns. Furthermore, it is also important to take leverage into account; asymmetric GARCH models produce the most reliable estimate for four out of six of the variables considered in this study. This is consistent with other findings in the literature.