Models

Question 1

Define: model

Answer 1

A model is a simplification of reality using financial, economical, math, and stats concepts

Question 2

Define: model risk

Answer 2

Risk of drawing incorrect conclusion due to limitations or flaws of the underlying model

Question 3

Identify considerations for severity of model failure
“How bad can it be?”

Answer 3

• Financial significance of item (B/S)
• Importance of model to the company
• Frequency of use of the model (sum to large total severity)
• Non-financial impact (Reputation risks)

Question 4

Identify considerations for likelihood of model failure

Answer 4

• Complexity of model
• Proper documentation on how to use the model
• Sufficiency of testing of the model
• Required knowledge and expertise of users
• adequacy of testing

Question 5

List the 3 elements of a model

Answer 5

Question 6

What is the element of a model: model specification

Answer 6

is the description of the components of a model and the interrelationship of those components with each other, including the types of data, assumptions, methods, entities, and events

Question 7

What is the element of a model: model implementation

Answer 7

the systems developed to perform the
calculations for a model specification (computer programs, spreadsheets, and database programs)

Question 8

What is the element of a model: model run

Answer 8

is a set of inputs and the corresponding results produced by a model implementation

Question 9

Model vs Calculation

Answer 9

The main distinction is the documentation required for a model (how it was chosen, how it’s used)

Question 10

Types of Models (4)

Answer 10

1) New model
2) Existing model (used in a new way)
3) Model approved by others
4) Model outside actuarial’s expertise

Question 11

Discuss validation for New Models (or substantially changes models)

Answer 11

4 Steps:

Question 12

What types of validation can be performed when using a model?

Answer 12

1) data (should be Reliable & Sufficient)
2) assumptions (some assumptions are not global but vary with model run)
3) results (should be reasonable relative to input)

Question 13

Explain term reliable and sufficient relative to model validation?

Answer 13

Reliable:
- data reconciles to other audited sources (Ex: balance sheet)
- data is reasonable with respect to prior period data (Ex: data from prior quarter shouldn’t be vastly different)

Sufficient:
- data fits model specification
- data is available in a consistent format