IFRS 9 Probability of Default Modeling Resources

Summary

The document is a short request for practical resources on modeling probability of default under IFRS 9, with an answer recommending two books on IFRS 9 and CECL credit risk modeling. It notes that the framework uses probability of default and loss given default, alongside other models.

The answer reports that one practical guide with worked examples in R and SAS helped its author understand the modeling work, and mentions a second book as a planned resource. It does not describe a PD estimation method, provide worked calculations, compare modeling approaches, or assess the recommended books in detail. Its value is therefore as a pointer toward learning materials rather than as a standalone guide to bank practice.

Key ideas

• The question seeks practical resources for IFRS 9 probability of default modeling.
• The answer identifies PD and LGD as components of the broader credit impairment modeling framework.
• A book with worked examples in R and SAS is cited as useful to the respondent.
• The document provides reading suggestions rather than a modeling method or evidence about bank practices.

Tags

# Probability of Default calculation


# Probability of Default calculation












I am looking for some good resources with handful of workout examples, on the modelling of the `Probability of Default` under `IFRS9` framework.

Could you please point me towards some good resources on this subject on how Banks typically calculate them?

Any insight will be very helpful.

## Answer by Dimitri Vulis (score 1)

https://quant.stackexchange.com/a/59244

IFRS 9 requires a bank to have a probability of defaut (PD) and a loss given defaut (LGD) and other models.

I looked at Tiziano Bellini IFRS 9 and CECL Credit Risk Modelling and Validation: A Practical Guide with Examples Worked in R and SAS and it helped me understand what's being done.

I also plan to look at Jing Zhang The New Impairment Model Under IFRS 9 and CECL when I get to it.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.