Chapter 6: Section 1.6

[jensen]

Hi

After question 6.8, the notes mentioned that the trigger event can be the modelled loss from a cat model after the model parameters have been updated to reflect the event.

Why is this so?

Thanks.

 

[Duncan Brydon]

For the benefit of other students reading this, I think you are referring to Section 1.6 of Chapter 7.

Modelled loss triggers may be preferred by investors because surprises are less likely to occur in the modelled losses than the actual losses. Investors may also be wary of actual loss triggers because the actual losses can be influenced by underwriting and claims handling decisions of the insurer sponsoring the bond.

I hope this helps
Duncan

 

[jensen]

Thanks Duncan

Oops! Typo there

What exactly do they mean by 'model parameters'? If it is an earthquake, is it referring to the magnitude/epicenter/depth? There are usually a couple of sources for these information following an event, so how do they select the parameters best represent the event?

 

[Duncan Brydon]

Yes, the types of item you suggest (magnitude etc) are the “model parameters”. A modelling agent would be assigned for the bond (eg AIR) at the outset and then the parameter values used for the event would be the ones collected by that modelling agent.

Hope this helps
Duncan