The article tests different risk models for two-asset portfolios using real data from the FTSE 100. They found that using Expected Shortfall along with Value-at-Risk is important for accurate risk estimation. The best model combined general Pareto marginals with Clayton copula, but it tended to be overly cautious in predicting losses. Overall, the study highlights the need for careful selection and testing of risk models for accurate risk assessment in financial portfolios.