GSE: Earnings Forecasting Framework Development
A $100+ billion government-sponsored enterprise with more than $3 trillion in assets sought to develop an end-to-end earnings forecast framework to project and stress-test the future performance of its loan portfolio. The comprehensive framework needed to draw data from a combination of unintegrated systems to compute earnings, capital management requirements and other ad hoc reporting under a variety of internal and regulatory (i.e., DFAST) stress scenarios.
Computing the required metrics required cross-functional team coordination, proper data governance, and a reliable audit trail, all of which were posing a challenge.
The Solution
RiskSpan addressed these needs via three interdependent workstreams:
Data Preparation
RiskSpan consolidated multiple data sources required by the earnings forecast framework. These included:
- Macroeconomic drivers, including interest rates and unemployment rate
- Book profile, including up-to-date snapshots of the portfolio’s performance data
- Modeling assumptions, including portfolio performance history and other asset characteristics
Model Simulation
Because the portfolio in question consisted principally of mortgage assets, RiskSpan incorporated more than 20 models into the framework, including (among others):
- Prepayment Model
- Default Model
- Delinquency Model
- Acquisition Model: Future loans
- Severity Model
- Cash Flow Model
Business Calculations and Reporting
Using the data and models above, RiskSpan incorporated the following outputs into the earnings forecast framework:
- Non-performing asset treatment
- When to charge-off delinquent loans
- Projected loan losses under FAS114/CECL
- Revenue Forecasts
- Capital Forecast
Client Benefits
The earnings forecast framework RiskSpan developed represented a significant improvement over the client’s previous system of disconnected data, unintegrated models, and error-prone workarounds. Benefits of the new system included:
- User Interface – Improved process for managing loan lifecycles and GUI-based process execution
- Data Lineage – Implemented necessary constraints to ensure forecasting processes are executed in sequence and are repeatable. Created a predefined, dynamic output lineage tree (UI-accessible) to build robust data flow sequence used to facilitate what-if scenario analysis.
- Run Management – Assigned a unique run ID to every execution to ensure individual users across the institution can track and reuse execution results
- Audit Trail – Designed logging of forecasting run details to trace attributes such as version changes (Version control system – GIT, SVN), timestamp, run owner, and inputs used (MySQL/Oracle Databases for logging)
- Identity Access Management – User IDs and access is now managed administratively. Metadata is captured via user actions through the framework for audit purposes. Role-based restrictions now ensure data and forecasting features are limited to only those who require such permissions
- Golden Configuration – Implemented execution-specific parameters passed to models during runtime. These parameters are stored, enabling any past model result to be reproduced if needed
- Data Masking – Encrypted personally identifiable information at-rest and in transit
- Data Management – Execution logs and model/report outputs are stored to the database and file systems
- Comprehensive User and Technical Documentation – RiskSpan created audit-ready documentation tied to logic changes and execution. This included source-to-target mapping documentation and enterprise-grade catalogs and data dictionaries. Documentation also included:
-
- Vision Document
- User Guides
- Testing Evidence
- Feature Traceability Matrix
-
