Operational risks in banks
Link to publisher version (URL)
Through empirical analysis of operational risk in a bank we derive a model to represent the distribution of losses, which we compare with more traditional models of operational risk. Our findings suggest that the generalised extreme value distribution provides a good fit to the annual loss distribution and that some conventional methods to model severity are inadequate because they neglect the extreme percentiles which are important in the type of analysis required under Basel II. Awarded a 2008 JASSA Merit award.