Risk aggregation and Extreme Events [78]

Go to: Summary | Previous | Next   
Bullet points include: In principle do not need to use Normal distribution as the ‘expected’ distribution C.f. definition of extreme event necessarily has in mind some prior view about what the distribution would be if it were not ‘fat-tailed’ In practice, Normal distribution is the most common But bear in mind need for distribution to have a finite standard deviation (and finite mean) to be able to standardise versus the Normal Not in general the case, e.g. all stable distributions other than the Normal distribution Need quite a few points to go ‘into the tail’

Contents | Prev | Next | ERM Lecture Series

Desktop view | Switch to Mobile