It states that due to non-linearity, a higher SCR could be produced for the 99.5% than that calculated by the correlation matrix.
Therefore doesn't this suggest the correlation matrix is insufficient since it also says an allowance has to be made for this, how? What kind of adjustment?
Does this non-linearity have any connection with least solvent likely event (LSLE) ? Does the difference between correlation matrix SCR and LSLE explain the allowance?
Final question on this- don't the SCR for each risk get calculated by a stochastic process anyway, so wouldn't that take care of the non-linearity?
Last edited by a moderator: Aug 26, 2016