What difference does it make to the VaR calculated in Example 19.2 if the exponentially weighted moving average model is used to assign weights to scenarios as described in Section 14.3?

Suppose that, in the example in Section 14.1, five stress scenarios are considered. They lead to losses (in $000s) of 235, 300, 450, 750, and 850. The probabilities assigned to the scenarios are 0.5%, 0.2%, 0.2%, 0.05%, and 0.05%, respectively. The total probability of the stress scenarios is, therefore, 1%. This means that the probability assigned to the scenarios generated by historical simulation is 99%. Assuming that equal weighting is used, each historical simulation scenario is assigned a probability of 0.99/500 = 0.00198. Table 14.4 is therefore replaced by Table 19.1. The probabilities assigned to scenarios are accumulated from the worst scenario to the best.⁸ The VaR level when the confidence level is 99% is the first loss for which the cumulative probability is greater than 0.01. In our example this is $282,204.