FRM Part II · FRM Exam Part II · Supervisory Guidance on Model Risk Management
A bank's VaR model at 99% confidence is backtested over 250 days and produces 9 exceptions. The validator compares this with the 2.5 expected exceptions and also finds that the exceptions cluster in a single stressed month. Which conclusion best reflects sound outcomes analysis and effective challenge?
Nine exceptions against 2.5 expected indicates the model underestimates risk, and clustering in a stressed month shows it does not respond to changing volatility. An effective validator would escalate, investigate root causes and require remediation or conservative adjustments, not accept the model.
- AThe model passes because the average daily loss was below VaR
- BThe exception count alone suffices; clustering is irrelevant to validation
- CThe model should be recalibrated only if exceptions fall below 2.5
- DThe count is well above expectation and clustering suggests the model fails to adapt to volatility, so the validator should escalate, investigate the cause and require remediation or conservative overlaysCorrect
Explanation
Expected exceptions are 250 x 1% = 2.5; 9 is more than three times that, indicating underestimation of risk. Clustering signals poor responsiveness to volatility regimes, so independent validators should escalate and require remediation. Average losses below VaR say nothing about tail exceptions.
Did you get it right without looking?
One question tells you little. A timed set on Supervisory Guidance on Model Risk Management shows your real accuracy, how long you take and where you lose marks.
More Supervisory Guidance on Model Risk Management questions
- Under supervisory guidance on model risk management (SR 11-7 style), which statement best describes the role of a bank's board of directors …
- A bank's model validation team reports to the head of the trading desk that relies on the models it validates, and its bonus is tied to desk…
- A bank developing a default model finds that the historical data for one key variable are sparse, so the developer substitutes a proxy serie…
- A bank licenses a credit scoring model from an external vendor. The vendor refuses to disclose source code, citing intellectual property. Un…
- A bank licenses a vendor-supplied market risk model whose code and methodology details are proprietary. How should the bank approach validat…
- A bank's internal audit reviews its model risk program. A senior business head has repeatedly overridden validators' findings on a high-mate…