Skip to main content
SpringerLink
Log in

IRB PD model accuracy validation in the presence of default correlation: a twin confidence interval approach

  • Original Article
  • Published:
Risk Management Aims and scope Submit manuscript

Abstract

The BIS indicated in July 2020 an unprecedented rise in default risk correlation as a result of pandemics-induced credit risks’ accumulation. A third of the world banking assets credit risk measurement depends on the Basel internal-ratings-based (IRB) models. To ensure financial stability, we wish IRB models to be accurate in default probability (PD) forecasting. There naturally arises a question of which model may be deemed accurate if the data demonstrates the presence of the default correlation. The existing prudential IRB validation guidelines suggest a confidence interval of up to 100 percentage points’ length for such a case. Such an interval is useless as any model and any PD forecast seem accurate. The novelty of this paper is the justification for the use of twin confidence intervals to validate PD model accuracy. Those intervals more concentrate around the two extremes (default and its absence), the higher the default correlation is.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Sources S&P SG is the time series of the speculatively graded companies having a Standards&Poors credit rating, see (S&P Global Ratings 2019, p. 3). Global GDP growth rate comes from the World Development Indicators database available at https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG

Fig. 2

Sources (S&P Global Ratings 2019, p. 3), authors’ calculations

Fig. 3
Fig. 4

Similar content being viewed by others

References

  • Aas, K., X.K. Dimakos, and A. Oksendal. 2007. Risk Capital Aggregation. Risk Management 9: 82–107.

    Article  Google Scholar 

  • Aivaizan, S., and V. Mkhitaryan. 1998. Applied Statistics and Basics of Econometrics. Moscow: Unity (in Russian).

    Google Scholar 

  • Altman, E. 2018. A Fifty-Year Retrospective on Credit Risk Models, the Altman Z-Score Family of Models and Their Applications to Financial Markets and Managerial Strategies. The Journal of Credit Risk, Volume IRMC 10th Anniversary Special Issue, pp. 1–34.

  • Altman, E.I. 1968. Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy. The Journal of Finance 23 (4): 589–609.

    Article  Google Scholar 

  • Aramonte, S. and F. Avalos. 2020. Corporate Credit Markets After the Initial Pandemic Shock. Available at: https://www.bis.org/publ/bisbull26.htm.

  • Baldwin, K., M. Alhalboni, and M.H. Helmi. 2019. A Structural Model of “alpha” for the Capital Adequacy Ratios of Islamic Banks. Journal of International Financial Markets, Institutions and Money 60: 267–283.

    Article  Google Scholar 

  • BCBS. 2005. Studies on the Validation on Internal Rating Systems. Available at: https://www.bis.org/publ/bcbs_wp14.htm [Accessed 25 December 2019].

  • Beaver, W. 1966. Financial Ratios as Predictors of Failure. Journal of Accounting Research 4: 71–111.

    Article  Google Scholar 

  • Blochwitz, S., M. Martin, and C. Wehn. 2006. III: Statistical Appoaches to PD Validation. In The Basel II Risk Parameters: Estimation, Validation, Stress-Testing—with Applications to Loan Risk Management, ed. B. Engelmann and R. Rauhmeier, 289–306. London: Springer.

    Chapter  Google Scholar 

  • Bluhm, C., L. Overbeck, and C. Wagner. 2003. An Introduction to Credit Risk Modeling, 1st ed. Boca Raton: Chapman & Hall.

    Google Scholar 

  • Brown, L., T. Cai, and A. DasGupta. 2001. Interval Estimation for a Binomial Proportion. Statistical Science 16 (2): 101–133.

    Article  Google Scholar 

  • Chang, Y.-P., and C.-T. Yu. 2014. Bayesian Confidence Intervals for Probability of Default and Asset Correlation of Portfolio Credit Risk. Computational Statistics 29 (1–2): 331–361.

    Article  Google Scholar 

  • Chiu, M., K. Jackson, and A. Kreinin. 2017. Correlated Multivariate Poisson Processes and Extreme Measures. Model Assisted Statistics and Applications 12: 369–385.

    Article  Google Scholar 

  • Duffie, D. and K. Singleton. 1999. Simulating Correlated Defaults. Available at: https://www.darrellduffie.com/uploads/working/DuffieSingleton1999.pdf [Accessed 15 December 2019].

  • EBA, 2019. Instructions for Reporting the Validation Results of Internal Models. IRB Pillar I Models for Credit Risk. Available at: https://www.bankingsupervision.europa.eu/banking/tasks/internal_models/shared/pdf/instructions_validation_reporting_credit_risk.en.pdf [Accessed 15 December 2019].

  • Genest, C., and J. Neslehova. 2007. A Primer on Copulas for Count Data. Austin Bulletin 37 (2): 475–515.

    Article  Google Scholar 

  • Hanson, S., and T. Schuermann. 2006. Confidence Intervals for Probabilities of Default. Journal of Banking & Finance 30: 2281–2301.

    Article  Google Scholar 

  • Höse, S. H. S., 2003. Simultaneous Confidence Intervals for Default Probabilities. In: Between Data Science and Applied Data Analysis. Studies in Classification, Data Analysis, and Knowledge Organization. Springer, New York

  • Karminsky, A. 2015. Credit Ratings and Their Modeling. Moscow: HSE Publishing House (in Russian).

    Google Scholar 

  • Kruppa, J., B. Lepenies, and K. Jung. 2018. A Genetic Algorithm for Simulating Correlated Binary Data from Biomedical Research. Computers in Biology and Medicine 92: 1–8.

    Article  Google Scholar 

  • Kumar, P.R., and V. Ravi. 2007. Bankruptcy Prediction in Banks and Firms via Statistical and intelLigent Techniques: A Review. European Journal of Operational Research 180 (1): 1–28.

    Article  Google Scholar 

  • Li, D. 2000. On Default Correlation: A Copula Function Approach. Journal of Fixed Income 9 (4): 43–54.

    Article  Google Scholar 

  • Li, X., Y. Shang, and Z. Su. 2015. Semiparametric Estimation of Default Probability: Evidence from the Prosper Online Credit Market. Economics Letters 127: 54–57.

    Article  Google Scholar 

  • Longin, F., and B. Solnik. 2001. Extreme Correlation of International Equity Markets. The Journal of Finance 6 (2): 649–676.

    Article  Google Scholar 

  • Lunn, A., and S. Davies. 1998. A note on generating correlated binary variables. Biometrika 2 (85): 487–490.

  • Nagpal, K., and R. Bahar. 2001. Measuring Default Correlation. Risk 14 (3): 129–132.

    Google Scholar 

  • Patel, K., and R. Pereira. 2008. The Determinants of Default Correlations. Advances in Econometrics 22: 123–158.

    Article  Google Scholar 

  • Penikas, H. 2020. Why the Conservative Basel III Portfolio Credit Risk Model Underestimates Losses? CEUR-WS 2795: 69–78.

    Google Scholar 

  • Preisser, J.S., and B.F. Qaqish. 2014. A Comparison of Methods for Simulating Correlated Binary Variables with Specified Marginal Means and Correlations. Journal of Statistical Computation and Simulation 84 (11): 2441–2452.

    Article  Google Scholar 

  • Qu, Y., P. Quan, M. Lei, and Y. Shi. 2019. Review of Bankruptcy Prediction Using Machine Learning and Deep Learning Techniques. Procedia Computer Science 162: 895–899.

    Article  Google Scholar 

  • S&P Global Ratings. 2019. 2018 Annual Global Corporate Default And Rating Transition Study. Available at: https://www.spratings.com/documents/20184/774196/2018AnnualGlobalCorporateDefaultAndRatingTransitionStudy.pdf. [Accessed 20 December 2019].

  • The Economist. 2020. A Sea of Debt. Corporate Bonds and Loans are at the Centre of a New Financial Scare. Available at: https://www.economist.com/finance-and-economics/2020/03/12/corporate-bonds-and-loans-are-at-the-centre-of-a-new-financial-scare

  • Vasicek, O. 2002. The Distribution of Loan Portfolio Value. Risk 12: 160–162.

    Google Scholar 

  • Witt, G. 2014. A Simple Distribution for the Sum of Correlated, Exchangeable Binary Data. Communications in Statistics Theory and Methods 43: 4265–4280.

    Article  Google Scholar 

  • World Bank. 2020. Pandemic, Recession: The Global Economy in Crisis. Available at: https://www.worldbank.org/en/publication/global-economic-prospects

  • Zaigraev, A., and S. Kaniovski. 2013. A Note on the Probability of at Least k Successes in n Correlated Binary Trials. Operations Research Letters 41 (1): 116–120.

    Article  Google Scholar 

Download references

Acknowledgements

Authors are grateful to the anonymous reviews whose comments helped to significantly improve the paper.

Funding

Funding was provided by National Research University Higher School of Economics (Basic Research Program).

Author information

Authors and Affiliations

  1. Department of Applied Economics, HSE University, Moscow, Russia

    Dmitriy Borzykh

  2. International Laboratory of Stochastic Analysis and Its Applications, National Research University Higher School of Economics, Moscow, Russia

    Dmitriy Borzykh

  3. Research and Forecasting Department, Bank of Russia, Moscow, Russia

    Henry Penikas

  4. Department of Applied Economics, International Centre for the Decision Choice and Analysis, National Research University Higher School of Economics, 11 Pokrovsky Boulevard, office T503, 109028, Moscow, Russia

    Henry Penikas

  5. P.N. Lebedev Physics Institute of the Russian Academy of Sciences, Moscow, Russian Federation

    Henry Penikas

Authors
  1. Dmitriy Borzykh
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Henry Penikas
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Henry Penikas.

Ethics declarations

Conflict of interest

On behalf of all authors, the Henry Penikas states that there is no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

See Table 1 in Appendix.

Table 1 Comparison of the default rate confidence intervals for the portfolio default rate \(\hat{p} = DR\)
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Borzykh, D., Penikas, H. IRB PD model accuracy validation in the presence of default correlation: a twin confidence interval approach. Risk Manag 23, 282–300 (2021). https://doi.org/10.1057/s41283-021-00079-2

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1057/s41283-021-00079-2

Keywords

JEL classification

Search

Navigation