Skip to content

FRM Part II · FRM Exam Part II · Non-parametric Approaches

A firm uses a 250-day historical simulation for 99% VaR. On day 251 the single worst loss in the window, a 9% fall that occurred on day 1, drops out of the sample and VaR falls sharply, though no market conditions have changed. Which description is most accurate, and what is the best remedy?

This is the ghost effect: an extreme old observation leaves the fixed window and VaR drops abruptly with no change in market conditions. Age-weighting or volatility-weighting gradually reduces old observations' influence, smoothing these jumps.

  1. AIt is a ghost effect; applying age-weighting or volatility-weighting reduces such abrupt jumpsCorrect
  2. BIt is a bias from assuming normality; switching to delta-normal removes it
  3. CIt is model risk from mapping; adding more risk factors removes it
  4. DIt is a backtesting exception; increasing the confidence level removes it

Explanation

Large observations that suddenly leave a fixed window cause abrupt VaR shifts, known as the ghost or shadow effect. Weighting schemes that decay the influence of old observations smooth the departure of old data. The other options misidentify the issue, as the method has no normality assumption and the jump is not a backtest exception.

Did you get it right without looking?

One question tells you little. A timed set on Non-parametric Approaches shows your real accuracy, how long you take and where you lose marks.

Practise Non-parametric ApproachesFree diagnostic mock

More Non-parametric Approaches questions