Analysis of the outputs, Risk measures, Risk appetite
The results thus consist often at the minimum in the projected balance sheets and P&L accounts for the time horizon chosen and according to the several thousands of scenarios created by the generator.
From these results, the probability distribution of the variable of interest - for example the accounting profit or loss - may be deduced.
To determine the economic capital - or another strategic variable, the use of the model being not limited to the determination of the economic capital - it remains to choose a risk measure and a confidence level or risk appetite.
A commonly chosen risk measure is the Value at Risk - or VaR - at a confidence level of say 99.5% (this is the risk measure and level of confidence to be used in the context of Solvency II, for example). In simple terms, the VaR at 99.5% of the accounting loss for example is the amount of loss for which only one scenario over 200 (ie 0.5%) gives a less favourable accounting loss. This risk measure has been criticised for different theoretical reasons, and it is why some insurers use also other risk measures, which also take into account the features of the scenarios for which this threshold of loss is overridden (with the VaR measure, these scenarios are only discarded) or multi-period risk measures.