Practical and theoretical challenges in validating Basel parameters: key learnings from the experience of a Canadian bank

Peter Miu and Bogie Ozdemir

Abstract

ABSTRACT

This paper, inspired by the efforts of a Canadian bank, discusses Basel preparation and validation issues. A comprehensive outcomes analysis (eg, back-testing) framework is presented, including a simulation-based calibration test. A consistent risk rating philosophy – point-in-time (PIT) or through-the-cycle (TTC) – encompasses both the probability of default (PD) and the default correlations, and the validation needs to be consistent with both. Related arguments are made that not only PD itself, but the correlation of PD used in economic capital models should be ratingsystem specific. We need to use a larger PD correlation under a TTC rating system than under a PIT rating system. Furthermore, Basel loss given default may not be appropriate for commonly used internal models, and accordingly adjustments are proposed.