Skip to content

FRM Part I · FRM Exam Part I · Measuring Return, Volatility, and Correlation

An analyst estimates correlation between two assets using 250 daily returns in a calm period, getting 0.30. In a later crisis, the assets frequently fall together. Which statement best explains why the calm-period estimate may understate crisis co-movement?

Correlation is unstable and tends to rise in stressed markets, so a coefficient estimated in a calm period can understate how strongly assets fall together in a crisis. A single linear measure also fails to capture tail dependence, making calm-period estimates unreliable for stress risk.

  1. ACorrelation is constant over time, so the crisis behavior is just sampling noise
  2. BCorrelation is typically unstable and often rises in stressed markets, so an unconditional estimate from calm data can understate joint tail riskCorrect
  3. CPearson correlation cannot be negative in calm markets, which biases the estimate downward
  4. DUsing daily rather than annual returns always produces a mechanically higher correlation, so 0.30 is overstated

Explanation

Correlations are not stable; they tend to increase during crises, and dependence in the tails can exceed what a single linear coefficient from calm data suggests. The other options assert false claims: correlation is not constant, can be negative, and the horizon claim is not a mechanical rule.

Did you get it right without looking?

One question tells you little. A timed set on Measuring Return, Volatility, and Correlation shows your real accuracy, how long you take and where you lose marks.

Practise Measuring Return, Volatility, and CorrelationFree diagnostic mock

More Measuring Return, Volatility, and Correlation questions