Skip to content

FRM Part II · FRM Exam Part II · Backtesting VaR

A bank's 99% VaR model produces 4 exceptions in 250 days, and the exceptions all occurred in the same two-week period. A validator applies only the Kupiec unconditional coverage test. What is the most appropriate conclusion?

The Kupiec test cannot detect clustering because it looks only at the count of exceptions, not their timing. Four exceptions is near the expected 2.5, so the model may pass, but an independence or conditional coverage test is needed to assess clustering.

  1. AThe test fully validates the model because 4 is close to the expected 2.5
  2. BThe test is inconclusive on clustering, since it ignores timing; an independence or conditional coverage test is also neededCorrect
  3. CThe test rejects the model because exceptions are clustered
  4. DThe test is invalid because the sample is under 500 days

Explanation

Kupiec considers only the number of exceptions, not when they occur. 4 versus an expected 2.5 is likely not rejected, yet clustering suggests the model fails to adapt to volatility changes. Christoffersen's independence or conditional coverage test addresses this.

Did you get it right without looking?

One question tells you little. A timed set on Backtesting VaR shows your real accuracy, how long you take and where you lose marks.

Practise Backtesting VaRFree diagnostic mock

More Backtesting VaR questions