Combining data sources to be used in quantitative operational risk models

Abstract

The management of financial losses is crucial for banks as they are required to set aside regulatory capital to absorb unexpected losses. Banks also need to calculate economic capital to ensure solvency according to their own risk profile. The main financial risks faced by banks are market, credit, and operational risk. Operational risk, the focus of the dissertation, includes fraud, improper business practices, regulatory risk, and others. Barings Bank’s loss of over USD1 billion due to rogue trading activities is well-known, but an extreme example of such risk. In order to calculate capital to withstand this risk, the aggregate distribution of operational losses for the next year is estimated. This distribution needs to be estimated in a forward looking manner and for this, assessments by experts are often used. The extreme quantiles of this distribution are of specific interest. For instance, a bank should hold capital to survive a one-in-a-thousand-year aggregate operational loss (the 99.9% Value at Risk of the distribution). A methodology is described to calculate capital for different operational risk categories. Banks often only have limited internal data available to accurately model the distribution and therefore use external sources and scenario assessments to supplement their data. Statistical methods are explored that could be used to combine limited historical data and scenario assessments provided by experts, to estimate the extreme quantiles of the aggregate distribution. This provides a way of constructing forward-looking distributions to calculate risk capital. SAS® OpRisk Global Data is used to demonstrate how external data can be used and scaled for use in the risk modelling process. Some measures are suggested that could be used to challenge experts to adjust their scenario assessments based on available historical data.

The main contribution of the research is to provide a holistic view of how internal data, external data and scenario assessments can be used to create a consistent framework for modelling operational risk capital within a bank or other financial institution.