This course prepares model implementation teams in financial institutions to conduct credit loss reserving and loan valuation to satisfy regulatory requirements, including IFRS9 and CECL.

Learn how to

• Import portfolio data and macroeconomic variables.
• Create and modify atomic models, including ASTORE and Python models, using SAS Risk Model Editor.
• Validate that a model is producing the same results as it did during the model estimation process.
• Write and apply logic to calculate specified output variables.
• Group models into model groups by asset type.
• View and analyze portfolio run results.
• Dynamically create shocked scenarios to evaluate the impact of changes in economic and portfolio variables on model results.
• Conduct attribution analyses to analyze the differences between two analysis runs by making incremental, sequential changes.
• Analyze current portfolio data and new volume projections.
• Publish models to a modeling system.

Who should attend

Members of model development and implementation teams in financial institutions who are responsible for activities such as stress testing, credit loss reserving, and loan valuation required to satisfy regulatory requirements

Formats available	Standard Duration
Classroom:	3 days
Live Web Classroom:	6 half-day session(s)	System Requirements
e-Learning:	13 hours/180 day license	System Requirements

Before attending this course, you should have experience using the SAS programming language. Expertise in statistical modeling concepts and methods is also required.

This course addresses SAS Model Implementation Platform software.

Getting Started with SAS Model Implementation Platform

• Introduction.
• Overview.

Creating Input Data Sets

• Overview of input data sets.
• Portfolio data set requirements.
• Economic data set requirements.
• Risk data object requirements.
• Function set requirements.
• Counterparty data set requirements.
• Mitigation data set requirements.

Working with Model Data Sets

• Model requirements.
• Implementing atomic models.
• Model unit testing.
• Model overrides data set requirements.

Creating Portfolio Analysis Objects

• Creating model groups and model group maps.
• Additional processing methods.
• Copying portfolio analysis objects.

Submitting and Analyzing Analysis Runs

• Submitting an analysis run.
• Viewing analysis results.
• Using modeling systems.
• Troubleshooting.

Additional Topics

• Working with ASTORE models.
• Working with Python models.
• Using model sensitivity analysis.
• Using mitigation.
• Creating a scenario run with backtesting.
• Performing attribution analyses.
• New originations generation.
• Cash flow analysis.