Establishing Model Risk Management
http://ccro.org © Copyright 2025, CCRO. All rights reserved. 6
2. Definitions
2.1. Model
Analytical models are sophisticated computational tools that organizations depend on to inform
financial decisions, optimize investments, shape operational strategies, manage risks, ensure
regulatory compliance, and support various other critical business functions and reporting.
They are typically imperfect approximations of reality, subject to the parameters and constraints
of their inputs and quantitative approaches.
SR 11-7 defines models as follows:
“The term model refers to a quantitative method, system or approach that applies
statistical, economic, financial, or mathematical theories, techniques and assumptions to
process input data into quantitative estimates…The definition of model also covers
quantitative as approaches whose inputs are partially or wholly qualitative based on
expert judgment, provided that the output is quantitative in nature.”
The flow and transformation of data within a model can be thought of as follows:
Figure 1: Flow and Transformation of Data
Not all calculations or processing approaches are considered “models” for this purpose. Rather,
the uncertainty attributable to assumption and theory selection, parameters and input choices,
and output interpretation is an essential determinant for something to be considered a model.
2.2. Model Risk
Model risk arises from actions taken or decisions made based on incorrect or misapplied model
outputs. The term “model risk” refers to the uncertainty that arises primarily due to:
• Misspecification: fundamental errors caused by improper development,
over/underfitting, or false validation that causes incorrect results
• Misuse: incorrect or inappropriate model usage, ignorance about limitations, and
other errors stemming from using a model without full understanding and
•Data: Quantitative,
Qualitative
•Assumptions
Input
•Formula,
Calculation,
Algorithm
•Theory,
Assumptions
Process
•Outcome:
Quantitative,
Qualitative
•Justification,
Interpretation
Output
http://ccro.org © Copyright 2025, CCRO. All rights reserved. 6
2. Definitions
2.1. Model
Analytical models are sophisticated computational tools that organizations depend on to inform
financial decisions, optimize investments, shape operational strategies, manage risks, ensure
regulatory compliance, and support various other critical business functions and reporting.
They are typically imperfect approximations of reality, subject to the parameters and constraints
of their inputs and quantitative approaches.
SR 11-7 defines models as follows:
“The term model refers to a quantitative method, system or approach that applies
statistical, economic, financial, or mathematical theories, techniques and assumptions to
process input data into quantitative estimates…The definition of model also covers
quantitative as approaches whose inputs are partially or wholly qualitative based on
expert judgment, provided that the output is quantitative in nature.”
The flow and transformation of data within a model can be thought of as follows:
Figure 1: Flow and Transformation of Data
Not all calculations or processing approaches are considered “models” for this purpose. Rather,
the uncertainty attributable to assumption and theory selection, parameters and input choices,
and output interpretation is an essential determinant for something to be considered a model.
2.2. Model Risk
Model risk arises from actions taken or decisions made based on incorrect or misapplied model
outputs. The term “model risk” refers to the uncertainty that arises primarily due to:
• Misspecification: fundamental errors caused by improper development,
over/underfitting, or false validation that causes incorrect results
• Misuse: incorrect or inappropriate model usage, ignorance about limitations, and
other errors stemming from using a model without full understanding and
•Data: Quantitative,
Qualitative
•Assumptions
Input
•Formula,
Calculation,
Algorithm
•Theory,
Assumptions
Process
•Outcome:
Quantitative,
Qualitative
•Justification,
Interpretation
Output
