Skip to content

FRM Part II · FRM Exam Part II · Beyond Exceedance-Based Backtesting of Value-at-Risk Models

A risk team backtests two 99% VaR models using only the count of exceedances over 500 days. Both models record 5 exceedances, yet one model's exceedances are large losses and the other's are barely above VaR. What is the main limitation this illustrates about exceedance-count backtests?

Exceedance-count tests ignore how large the losses beyond VaR are. Two models with identical exceedance counts but very different tail severity look equally good, which is why scoring functions that use loss magnitude are used to compare and rank VaR models.

  1. AThey ignore the magnitude of the losses beyond VaR, so they cannot separate the modelsCorrect
  2. BThey require the VaR to be computed using historical simulation only
  3. CThey cannot be applied to 99% confidence levels
  4. DThey always reject models with more than zero exceedances

Explanation

A count-based test treats every exceedance as a simple 0/1 event. It carries no information on how large the loss beyond VaR was, so two models with the same count but different tail severity look identical. Scoring-function approaches use loss size to discriminate.

Did you get it right without looking?

One question tells you little. A timed set on Beyond Exceedance-Based Backtesting of Value-at-Risk Models shows your real accuracy, how long you take and where you lose marks.

Practise Beyond Exceedance-Based Backtesting of Value-at-Risk ModelsFree diagnostic mock

More Beyond Exceedance-Based Backtesting of Value-at-Risk Models questions