The Credit Risk Modelling specialist is responsible for the development, oversight and embedding of credit risk measurement models for the Company.
The Credit Risk Model specialist plays an important role to develop the credit risk models to predict risk estimates such as PD, EAD, and LGD, and operational models to support credit risk decisions.
Model Development and Validation:
- Develop, implement, and maintain Probability of Default (PD), Loss Given Default (LGD), and Exposure at Default (EAD) models for assessing credit risk.
- Develop and validate macro-economic linked models for stress testing and incorporate machine learning models (e.g., random forests, gradient boosting, neural networks) as appropriate.
- Develop fraud detection models, leveraging supervised and unsupervised machine learning techniques (e.g., anomaly detection, clustering) to identify potential fraud indicators.
- Work with a variety of credit risk models, including structural models (e.g., Merton, KMV), reduced-form models (e.g., Jarrow-Turnbull), scoring models (e.g., logistic regression, Altman's Z-Score), and transition matrix models (e.g., CreditMetrics).
- Utilize coding languages, including Python, R, SAS, and SQL, for data processing, statistical modeling, and automation of model development.
- Conduct in-depth validation and back-testing of models to ensure accuracy and compliance with regulatory requirements.
- Document and present model development processes, assumptions, and limitations for audit and regulatory review.
Data Analysis:
- Analyze large datasets from multiple sources to extract insights that inform model development, credit risk assessment, and fraud detection.
- Ensure data quality and consistency across modeling processes.
- Perform exploratory data analysis using tools like Pandas and NumPy (Python), Tidyverse (R), and SQL.
Risk Assessment:
- Work closely with credit risk management and fraud detection teams to understand business needs and tailor models to align with the company risk and fraud detection strategies.
- Conduct scenario analysis and stress testing to assess potential risks under different economic and fraud scenarios.
- Collaborate with business stakeholders to incorporate model insights into decision-making processes for credit and fraud risk management.
- Monitor emerging fraud patterns and update fraud detection models to address new fraud schemes and trends.
Continuous Improvement:
- Monitor model performance and make adjustments as needed to improve accuracy and productiveness.
- Stay updated on industry best practices, new statistical techniques, and changes in regulatory guidance.
Develop tools and scripts to automate modeling processes and improve efficiency using Python and SQL.
