Model Validation
Finalyse’s ‘Model Validation’ service determines whether the models supporting your business decisions and the monitoring framework that assures their reliability are methodologically sound and compliant with relevant regulations such as Basel III or your internal standards. In addition, this service provides you with a comparison to industry best practices. The resulting ‘Validation Report’ gives you an independent and detailed assessment of your models and the related infrastructure. It identifies the strong and weak points of your models, their fit to your environment and actions for improvement.
You will benefit from the sound track record of Finalyse of more than 20 successful projects in regulatory and internal compliance gap analysis, qualitative and quantitative analysis of IRB & IFRS9 models (incl. existing scorecards, PD, LGD, EAD/CF), IRRBB, Market Risk (FRTB, VaR, Stress Testing), Fraud, Machine Learning and other models.
How does Finalyse address your challenges?
Identification of major model weaknesses and enhancement possibilities
Identification of gaps vs. regulatory/internal requirements PLUS ready-made materials for regulators and/or internal audit
Mitigation of model risk management
Getting best practice benchmarks
Placing the bank’s models within the model landscape of the industry
Full validation cycle from model objective to use test
Key Features
- Our ‘Model Validation’ service offers a second opinion on your models supporting business decisions to ensure their methodological soundness and compliance with regulations and your internal standards.
- As a result, we will provide you with a ‘Validation Report’, an independent assessment of your models and their related implementation aspects(data, IT, etc.).
What Finalyse experts say
This article introduces the challenges of the ECB supervised entities regarding their internal validation of the IRB Approach. It also considers a benefit of use of Model Validation tool.
ReadMachine learning is today’s buzzword and it has gone through some phenomenal changes over the last few years. However, despite widespread adoption, machine learning models remain mostly black boxes. Hence it is essential to have techniques for explaining these black boxes in an interpretable manner. This article looks at the methods that are most used for interpreting machine learning models: SHAP, LIME and CP Profiles. It discusses advantages and drawbacks of each and shows how are they being used in practice.
Read