Estimating VaR and ES of EU ETS based on GARCH-type models with time-varying higher order moments

The focus of this study is on the risk measurement performance of the European Union Emissions Trading System (EU ETS) using GARCH-type models with and without time-varying higher order moments. We propose the GJR-GARCHSK model, which incorporates asymmetric volatility into the higher order moments. The empirical results show that the time-varying conditional variance, skewness, kurtosis and asymmetry parameters are all statistically significant. The time-varying higher order moment volatility models perform better for both VaR and ES and are robust to different sample periods. In many cases, the GJR-GARCHSK model presents a high degree of risk measurement accuracy, and thus it provides a new approach for future research into financial risk management.