FRM Part II · FRM Exam Part II · Parametric Approaches (II): Extreme Value
A risk manager uses the GEV distribution for block maxima of daily losses with location μ = 2.0%, scale σ = 1.0% and shape ξ = 0 (Gumbel). Blocks are 20 trading days. The cumulative distribution is F(x) = exp(-exp(-(x-μ)/σ)). What is the loss x such that the block maximum exceeds it with probability 1%? (Use ln(-ln 0.99) = -4.600.)
The 99th percentile of the Gumbel block maximum is μ plus σ times 4.600, which gives 2.0% + 4.6% = 6.6%. This is the loss exceeded by the 20-day block maximum with 1% probability.
- A4.0%
- B6.6%Correct
- C2.0%
- D9.2%
Explanation
Set F(x)=0.99: -(x-μ)/σ = ln(-ln 0.99) = -4.600, so x = 2.0% + 1.0% × 4.600 = 6.6%. The 4.0% option wrongly drops the factor and uses 2 times σ; 9.2% doubles the 4.6 term.
Did you get it right without looking?
One question tells you little. A timed set on Parametric Approaches (II): Extreme Value shows your real accuracy, how long you take and where you lose marks.
More Parametric Approaches (II): Extreme Value questions
- A risk analyst models the extreme daily losses of an equity portfolio whose returns are well described by a Student-t distribution with 4 de…
- An analyst studies the largest daily loss in each of many 20-day blocks for a trading portfolio. According to the Fisher-Tippett-Gnedenko th…
- A bank's daily losses are modeled using a peaks-over-threshold approach. The threshold u is 2.0 (in millions), 5% of observations exceed it,…
- An analyst plots the sample mean excess function of portfolio losses against the threshold. Above a loss of 3% the plot is approximately lin…
- A risk analyst wants to estimate the probability that two equity indices will both suffer extreme losses in the same period. Which multivari…
- A risk analyst fits a generalized extreme value (GEV) distribution to the monthly maximum daily losses of a trading portfolio. The estimated…