Establishing Model Risk Management
http://ccro.org © Copyright 2025, CCRO. All rights reserved. 9
2. Development: The model request and development processes, including selection and
design development implementation, testing, performance monitoring planning, and
documentation.
3. Validation: The independent verification of assumptions, choices, operability, and
results from the model development process. Validation incorporates effective challenge
principles and typically includes analysis, testing and challenger models,
documentation, risk tiering, and approval processes to move the model from
development to implementation.
4. Implementation: The implementation and deployment of models following proper
validation and approval. Implementation processes often include change management
procedures, upstream and downstream data dependency identification, communication
and training plan rollout, and prior-model restoration plans if the implementation fails.
5. Usage: The use of models for their intended purposes. Usage processes include
receiving training, supervision of access rights and other controls, running the model
and accessing its outputs.
6. Ongoing Performance Monitoring: Statistical performance reviews are carried out on a
regular basis. Monitoring processes often include escalations when results deviate from
stated use cases or predefined tolerance parameters. Back-testing processes often
include parameterization reviews and recalibration as necessary to maintain model
accuracy and usefulness.
7. Periodic Review &Revalidation: The post-validation review and revalidation are
carried out by the validation team on a regular basis.
http://ccro.org © Copyright 2025, CCRO. All rights reserved. 9
2. Development: The model request and development processes, including selection and
design development implementation, testing, performance monitoring planning, and
documentation.
3. Validation: The independent verification of assumptions, choices, operability, and
results from the model development process. Validation incorporates effective challenge
principles and typically includes analysis, testing and challenger models,
documentation, risk tiering, and approval processes to move the model from
development to implementation.
4. Implementation: The implementation and deployment of models following proper
validation and approval. Implementation processes often include change management
procedures, upstream and downstream data dependency identification, communication
and training plan rollout, and prior-model restoration plans if the implementation fails.
5. Usage: The use of models for their intended purposes. Usage processes include
receiving training, supervision of access rights and other controls, running the model
and accessing its outputs.
6. Ongoing Performance Monitoring: Statistical performance reviews are carried out on a
regular basis. Monitoring processes often include escalations when results deviate from
stated use cases or predefined tolerance parameters. Back-testing processes often
include parameterization reviews and recalibration as necessary to maintain model
accuracy and usefulness.
7. Periodic Review &Revalidation: The post-validation review and revalidation are
carried out by the validation team on a regular basis.
