Shivam Vats

Client - Discover Financial Services (DFS)

Model Validation of Probability of Default (PD) models for Bank's retail portfolio. The project aimed to ensure the accuracy and reliability of the PD model used for assessing credit risk in our retail portfolio.

Key Task Performed :-

• Involved in quarterly model monitoring and validation of PD models.
• Involved in evaluating effectiveness of current policies by tracking them on monthly basis through excel reports and on annual basis through Annual Model Review (AMR) and also involved in revalidation of Model.
• In case of breach, perform root cause analysis.
• Model Documentation following MRM (Model Risk Management) guidelines.
• Performance Matrices Used - KS Statistics, PSI, Bad Rate, Precision, Recall, and Sensitivity.
• Sounds knowledge of IFRS9 provisions.
• Good knowledge of IFRS9 compliance, demonstrating a deep understanding of impairment modeling and forward-looking provisioning.
• Strong conceptual understanding of various functional/technical skills.
• Strong analytical and credit risk background, including the ability to apply standard statistical techniques using SAS to extract and analyze data.
• Leveraged SAS language for data extraction, transformation, and loading (ETL) to prepare and preprocess the dataset for model validation.
• Ensured the thorough documentation of all validation procedures and results, including SAS code and methodology.

• Part of Mortgage - Data Reporting & Analytics Team.
• Responsible for updating/executing codes as per the requirement by stakeholders.
• Involved in performing impact analysis after the code change.
• Involved in data extraction and analysis using SQL.
• Looking into adhoc data requests coming from business.
• Able to work directly with Operations and Business team to ensure results meet their needs.