A practical guide to measure operational risk using subjective data through copulas and scenario analysis

Marco Folpmers

Abstract

ABSTRACT

In this article we give a step-by-step illustration of how to use copulas to measure operational risk within the loss distribution approach (LDA). The advantage of this copula approach is that it takes correlated operational risk events into account. In our example, the operational value-at-risk is calculated using a simulation based on correlated Poisson frequency distributions and triangularly distributed damages. The parameters need to be established with the help of loss data and expert panels. These expert panels are also needed for an assessment of the correlations. These correlations could also be based on proxy data, eg, correlations calculated with the help of time series data of key risk indicators.