Establishing Model Risk Management
http://ccro.org © Copyright 2025, CCRO. All rights reserved. 15
5.4. Framework Delivery
The MRM framework should be used to determine what should be designed and implemented
as part of the MRM program rollout. A sound MRM program should have standards
addressing analytics, model development, and documentation as noted in 4.2. Formalized
processes and approvals for implementing, testing, validating, usage, and monitoring should
accompany the company’s MRM standards. When designing the program, companies should
begin with an initial current state assessment and consider structured processes for model
approval and deployment, backtesting, deficiency tracking, and remediation. Final delivery
should include appropriate stakeholder engagement, change management, and staff education
to ensure meaningful adoption.
Recommendations:
1. Determine the standards, processes, approvals, and other controls to be implemented,
assess the company’s current state and readiness, and develop a plan to implement an
MRM program consistent with the company’s risk profile.
2. Define a clear, actionable structure for tolerance threshold exceptions during validation
and approval processes.
3. Maintain centralized templates and documentation guidance at all model life cycle
stages.
4. Engage stakeholders in the design and delivery of the MRM program, and employ
change management principles to ensure a successful rollout.
5.5. Inventory
Model inventories provide consistent identification and storage of models and analytical
methods. The inventory process should involve identifying functional capabilities delivered to
the organization via analytics and models. This includes vendor models, internally developed
models, spreadsheets and EUCs, outsourced analytics and models, and legacy modeling
processes. Model inventories should be supported by technology to enable secure storage and
management of the inventory and related workflows.
Recommendations:
1. Model inventories should catalog models at all stages in the model lifecycle, including
categories for models that are in development, deployed, and retired.
2. Inventories should capture relevant modeling metadata such as model identification,
purpose, risk assessment, and business function alignment.
3. Inventory systems should store or link to current model documentation, including
model development, validation, and implementation materials.
http://ccro.org © Copyright 2025, CCRO. All rights reserved. 15
5.4. Framework Delivery
The MRM framework should be used to determine what should be designed and implemented
as part of the MRM program rollout. A sound MRM program should have standards
addressing analytics, model development, and documentation as noted in 4.2. Formalized
processes and approvals for implementing, testing, validating, usage, and monitoring should
accompany the company’s MRM standards. When designing the program, companies should
begin with an initial current state assessment and consider structured processes for model
approval and deployment, backtesting, deficiency tracking, and remediation. Final delivery
should include appropriate stakeholder engagement, change management, and staff education
to ensure meaningful adoption.
Recommendations:
1. Determine the standards, processes, approvals, and other controls to be implemented,
assess the company’s current state and readiness, and develop a plan to implement an
MRM program consistent with the company’s risk profile.
2. Define a clear, actionable structure for tolerance threshold exceptions during validation
and approval processes.
3. Maintain centralized templates and documentation guidance at all model life cycle
stages.
4. Engage stakeholders in the design and delivery of the MRM program, and employ
change management principles to ensure a successful rollout.
5.5. Inventory
Model inventories provide consistent identification and storage of models and analytical
methods. The inventory process should involve identifying functional capabilities delivered to
the organization via analytics and models. This includes vendor models, internally developed
models, spreadsheets and EUCs, outsourced analytics and models, and legacy modeling
processes. Model inventories should be supported by technology to enable secure storage and
management of the inventory and related workflows.
Recommendations:
1. Model inventories should catalog models at all stages in the model lifecycle, including
categories for models that are in development, deployed, and retired.
2. Inventories should capture relevant modeling metadata such as model identification,
purpose, risk assessment, and business function alignment.
3. Inventory systems should store or link to current model documentation, including
model development, validation, and implementation materials.
