-
Methods, analytics and models used for: pricing, provisioning (IFRS 9 and CECL), stress-testing (EU and CCAR/DFAST) and economic/regulatory capital
-
Risk parameters: Probability of Default (PD), Exposure at Default (EAD/CCF/LEQ), Loss Given Default (LGD), Correlation and Dilution modelling
-
Portfolio models: loss distribution and securitisation modelling. Structured products’ pricing. Empirical calibration of assets' correlations (Vasicek framework)
-
For the asset classes: Securitization, Corporates (Large and SMEs), Banks and FIs, Sovereigns/Municipalities/PSEs, Specialized Lending (Aircraft, Shipping, Project, Commodity and Real Estate Finance), Private Banking, Retail, and Purchased Receivables
-
Trading book: adjustments to derivative contracts' valuations (CVA)
-
Development or independent-validation of models
We support our clients on the following topics:
REGULATION
Our regulation expertise encompasses a comprehensive understanding of the ever-evolving regulatory landscape. We navigate the complexities of compliance to ensure that clients’ businesses operate seamlessly within the bounds of financial industry regulations. Our team stays abreast of regulatory changes, offering strategic insights and tailored solutions.
DATA
At the heart of our consulting firm's operations is a dynamic team of adept data professionals. Before entering any modelisation process, our team knows how to build comprehensive and accurate datasets. This meticulous process lays the groundwork for robust analyses, setting the stage for modelisation and/or informed decision-making.
MODELLING
Our consulting firm gathers a skilled team of statistical engineers who apply cutting-edge methodologies to address credit risk. Our approach is anchored in the utilization of methodologies compliant with regulatory requirements, internal standards and industry's best practices.
Use Case
IRB-repair of a Financial-Institutions rating system
Context and scope:
As part of an IRB-repair program, Neon Risk has been selected by a top-tier French banking group (first line of defense modelling team) to remediate its financial institutions rating system. This rating system covers counterparties such as: banks, insurers, funds, and other non-bank financial institutions (incl. independent asset managers and brokers).
Challenges:
This banking/trading book segment is a Low Default Portfolio (LDP), necessitating the use of specific statistical methods both for the risk differentiation and quantification. Expert judgment is very key in such project. And benchmarking capabilities are required for initial validation.
Success Factors:
-
Strong implication of sponsors
-
Clear scoping of expectations and roadmap with sufficient buffers
-
Early identification of core group of relevant stakeholders and experts
-
Specification of historical datasets with experts, post inventory of IRB-repair requirements
-
Development of a complete statistical workflow enabling data visualization, coverage and representativeness analyses, multinomial regression (and machine learning challenger models), combinatory analyses to retain the most discriminant risk drivers (cf. cost of use and maintenance), performance and stability tests
-
A strong benchmarking capability (towards rating agencies and dedicated benchmarking solutions)
-
Continuous, constructive and transparent information of the independent validation team (to anticipate red flags)
This project highlights our dedication to advancing credit risk modelling practices within the framework of regulatory compliance, ensuring performance and cost-effectiveness of our models.