Measuring Portfolio Credit Risk: Modelling Versus Calibration Errors

14 Pages Posted: 23 Jun 2012 Last revised: 29 Sep 2013

See all articles by Nikola A. Tarashev

Nikola A. Tarashev

Bank for International Settlements (BIS) - Monetary and Economic Department

Haibin Zhu

Bank for International Settlements (BIS)

Date Written: March 1, 2007

Abstract

A model-based assessment of credit risk is subject to both specification and calibration errors. Focusing on a well known credit risk model, we propose a methodology for quantifying the relative importance of alternative sources of such errors and apply this methodology to a large data set. We find that flawed calibration of the model can substantially affect the measured level of portfolio credit risk. By contrast, a model mis-specification generally has a limited impact, especially for large, well diversified portfolios.

JEL Classification: C15, G13, G21, G28

Suggested Citation

Tarashev, Nikola A. and Zhu, Haibin, Measuring Portfolio Credit Risk: Modelling Versus Calibration Errors (March 1, 2007). BIS Quarterly Review, March 2007, Available at SSRN: https://ssrn.com/abstract=1967437

Nikola A. Tarashev (Contact Author)

Bank for International Settlements (BIS) - Monetary and Economic Department ( email )

Centralbahnplatz 2
CH-4002 Basel
Switzerland

Haibin Zhu

Bank for International Settlements (BIS) ( email )

Hong Kong
Hong Kong
852 2878 7145 (Phone)
852 2878 7123 (Fax)

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
80
Abstract Views
557
Rank
551,501
PlumX Metrics