The purpose of this workshop is to give you a good, "hands-on" understanding of methods and tools for measuring and managing of operational risk.
We will start with a brief review of the different types of operational risk events and their loss impacts.
We then explain how operational risk can be modeled using internal and external loss data, self-assessments and other techniques. We discuss the problem in collecting and validating relevant (and sufficient) data for reliable estimates of a loss distribution. We give examples of how data for loss frequency by business line and event type can be obtained from external loss databases such as ORX and we demonstrate how this data can be combined with internal data and qualitative assessments to construct a loss distribution and to calculate expected and unexpected losses. We explain and demonstrate how the loss distribution and its associated parameters can be used to calculate regulatory and economic capital.
Further, we explain, discuss and demonstrate how to verify and validate the OP risk framework and risk models. Methods include back testing and statistical testing. Specifically, we explain why the difficulties in conducting back testing owing to data availability and we show how it may be possible to secure relatively robustness of operational risk measurement with statistical testing.
We also explain and demonstrate how to deal with high severity/low frequency events and we discuss how correlations between different OP risk events can be incorporated, and how scenario analysis may be used to examine sensitivity of the estimation to changes in the assumptions in order to check the robustness of models in stress situations.
Finally, we discuss practical issues related to management, control and reporting of OP risk.
