Combining underreported internal and external data for operational risk measurement

Montserrat Guillen, Jim Gustafsson, Jens Perch Nielsen

Abstract

ABSTRACT

Operational risk data sets have two types of sample selection problem: truncation below a given threshold owing to data that is not recorded and random censoring above that level caused by data that is not reported. In this paper we propose a model for operational losses that improves the internal loss distribution modeling by combining internal and external operational risk data. We also consider the possibility that internal and external data has been collected with a different truncation threshold. Moreover, the model is able to cope with unreported losses by means of an estimated underreporting function.