Non-linearity and non-seperability of risks

How do companies allow for this in their calculation of SCR? The core reading mentions that it has to be allowed for but does not specify how, when etc.

 

Standard formula does not explicitly allow for non-linearity. Standard formula allows for non-separability to a degree when aggregating risks using correlation matrices.

Internal model firms will have flexibility in how they address non-linearity and non-separability, and approaches will vary between firms. I suspect [but cannot say for certain] that details of how its done is outside scope of core reading/exam. If you are interested in learning more, try searching for and reading up on the "Equivalent Scenario" or on "Copulas".