Multivariate approaches for Value-At-Risk and Expected Shortfall on electricity forwards

This study applies a group of multivariate volatility models to forecast 1-day ahead electricity forward returns with data supplied by Swedish energy company Skelleftea Kraft AB. From these we estimate the risk measures Value-at-Risk and Expected Shortfall. We find that there is seasonality in volatility which results in four distinct portfolios for analysis. Four multivariate volatility models are applied to these portfolios which are then subject to minimum variance optimization. The forecasted risk measures of the portfolios is then treated to various backtesting procedures to assess which model is most suitable. The models are able to predict well but we find no strong results that favors any particular model in the backtests, although some indications lead us to recommend the industry standard RiskmetricsTM model.