FRM Part II · FRM Exam Part II · Estimating Market Risk Measures: An Introduction and Overview
A risk committee considers a spectral risk measure that places weights on the quantiles of the loss distribution. Which weighting function would make the measure coherent and reflect risk aversion?
A spectral risk measure is coherent with risk aversion when its weights are non-negative, sum to one and rise with loss severity, so worse outcomes receive more weight. A single-quantile weight gives VaR, which is not coherent, and declining weights contradict risk aversion.
- AWeights that are non-negative, sum to one and increase with the severity of the loss quantileCorrect
- BEqual weights on all quantiles, which gives the expected loss
- CZero weight on all quantiles except the 95th, which gives VaR
- DWeights that decline as losses become more severe
Explanation
A spectral measure is coherent when the weights are non-negative, integrate to one and are non-decreasing in loss severity, which expresses risk aversion. Equal weights give the mean, a single-quantile weight gives VaR (not coherent), and declining weights contradict risk aversion and break coherence.
Did you get it right without looking?
One question tells you little. A timed set on Estimating Market Risk Measures: An Introduction and Overview shows your real accuracy, how long you take and where you lose marks.
More Estimating Market Risk Measures: An Introduction and Overview questions
- A risk manager notes that an exponential spectral risk measure becomes much larger when the risk-aversion coefficient is increased. What is …
- A portfolio manager models daily portfolio P/L as normally distributed with a mean of USD 0 and a standard deviation of USD 2 million. Using…
- A risk analyst wants a standard error and confidence interval for a 99% expected shortfall estimated from 750 historical returns, without as…
- A portfolio manager models daily portfolio P/L as normally distributed with a mean of USD 20,000 and a standard deviation of USD 100,000. Us…
- A risk manager computes a spectral risk measure on ten equally likely sorted losses (in $ millions): 2, 3, 4, 5, 6, 7, 8, 9, 10, 20. Weights…
- A risk manager compares the 99% 1-day VaR from historical simulation using 1,000 days of data against a parametric normal VaR. Which stateme…