Model risk
The risk of loss because a model is wrong or wrongly used: a mistaken distribution, bad parameters, or output trusted without checking. More simulation trials do not reduce it; validation and oversight do.
Where you will meet it
Model risk appears in 4 learning outcomes on the CFA syllabus. A few of them:
- Contrast historical simulation with Monte Carlo simulation and with bootstrap resampling
“Monte Carlo simulation draws from probability distributions the analyst specifies, which allows outcomes more extreme than any observed and makes it easy to test alternative assumptions, but exposes the results to model risk if the distribution or its parameters are wrong.”
- Contrast discretionary fundamental management with quantitative active management
“Their typical weaknesses differ too: behavioural bias for fundamental managers, model risk and data-mining for quants.”
- Explain where AI and large language models fall short or pose risks in investment work
“Together these form model risk, and the standard response is validation, documented controls and human oversight of every consequential output.”
- Explain what can go wrong when a portfolio is managed against a set of liabilities
“Model risk arises because liability values rest on assumptions: for a pension, the discount rate, salary growth, inflation and longevity.”
Practise it
Practise model risk until it is automatic: questions that explain every wrong option.
Take the free diagnostic →