Skip to content

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

A risk manager reviews a bank's 99% one-day VaR model and finds that over 250 days there were 2 exceptions, which falls inside the regulatory green zone. A colleague argues that this alone proves the model is accurate. Which limitation of exceedance-based backtesting most directly undermines that conclusion?

The key limitation is that exceedance counting ignores the magnitude of losses beyond VaR. A model with few exceptions may still understate tail severity, so passing the count test does not prove accuracy of the model's tail behavior.

  1. AException counts ignore the size of losses beyond VaR, so the model could badly understate tail severityCorrect
  2. BException counts can only be computed for 95% confidence levels
  3. CException counts require the P&L distribution to be normal
  4. DException counts cannot be used when the portfolio contains options

Explanation

An exceedance test records only whether the loss exceeded VaR, not by how much. A model can show few exceptions yet have very large losses when they occur. The other options misstate the requirements of the test; it works at any confidence level, does not need normality, and can be applied to any portfolio.

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