Der Vortrag „Challenges and Pitfalls in measuring operational risk“ von Edu Pristine ist Bestandteil des Kurses „Archiv - Operational Risk“. Der Vortrag ist dabei in folgende Kapitel unterteilt:
5 Sterne |
|
5 |
4 Sterne |
|
0 |
3 Sterne |
|
0 |
2 Sterne |
|
0 |
1 Stern |
|
0 |
... vary from quite small and predictable to large and unexpected. Operational losses are described as heavy-tailed data. Traditional mean and standard deviation concepts ...
... total loss is largely driven by relatively few observation in the data set Dominance of mixture: The heavy tailed distribution ...
... less precision relative to well behaved distributions, such as the normal distribution. Therefore larger sample sizes are needed to estimate high quantile losses with reasonable bounds. ...
... several underlying loss generating mechanism, but some are more likely to yield extreme events. The extreme events are not drawn from a known distribution and/or do not offer a pattern for estimation. Generalized Pareto Distribution (GPD) can also be used to model extreme events. GPD are flexible ...
... mandated by Basel Committee to estimate the annual loss for the bank. Assume each loss distribution to be independent and identically distributed. Annual losses for unit of ...
... Basel mandate. To fulfill this mandate, we need 1000 year bank data. Further, even if the data is available, there ...