The EBA recently released its much-anticipated follow-up report to its 2021 consultation on using machine learning (ML) for IRB models. The principle-based recommendations remain largely unchanged and are welcomed by the industry. We summarise the key points of the report including what firms are using or intending to use ML for, as well as the need for firms to consider other legal frameworks when using ML. Finally, we conclude with our own approach to developing compliant "synthetic" ML models for prudential purposes and how they can be used to overcome some of the challenges of using ML.

Read our published briefing here.

If you missed our previous posts, please follow the links below:
Part I - A brief overview of the regulatory landscape
Part II - The foundations of machine learning
Part III - The power of machine learning

For more information or if you have any questions please contact:

Dr. Andreas Peter
Managing Partner
Fintegral Group
+49 160 583 40 66
andreas.peter@fintegral.com

Dr. Tobias Kesselring
Senior Manager
Fintegral Schweiz AG
+41 79 271 19 00
tobias.kesselring@fintegral.com

Abdul Qaiyum
Senior Consultant
Fintegral UK Ltd.
+44 7496 363 298
abdul.qaiyum@fintegral.com