Skip to content

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

A bank wants to compare VaR models using a method that rewards models for the size of tail losses rather than only the frequency of breaches. Which approach best fits this aim?

Averaging a consistent quantile scoring function over the sample and comparing across models best rewards tail-loss magnitude. Count-based tools such as the traffic-light test ignore severity, and choosing the lowest VaR simply rewards understating risk.

  1. AAverage a consistent quantile scoring function over the sample and compare across modelsCorrect
  2. BCount exceedances and apply the traffic-light test
  3. CCompare only the unconditional coverage p-values
  4. DRank models by the lowest average VaR figure

Explanation

A quantile scoring function uses the size of the loss relative to VaR, so it reflects tail severity as well as frequency. Traffic-light and coverage tests use only counts. Ranking by lowest VaR would reward under-reporting risk, not accuracy.

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