Skip to content

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

An analyst computes the correlation between the daily returns of two assets using a window containing a few extreme, same-direction outlier days. Compared with the correlation over a calm period, which is the most likely effect?

The estimated correlation is likely overstated. Sample correlation uses products of deviations from the mean, so a few extreme days where both assets move strongly in the same direction dominate the calculation. Being scale-free does not make correlation robust to outliers.

  1. AThe estimated correlation is overstated, because a few shared extreme observations dominate the co-movementCorrect
  2. BThe estimated correlation falls toward zero, because outliers add noise
  3. CThe correlation is unaffected, because correlation is scale-free
  4. DThe correlation becomes negative, because outliers reverse the sign

Explanation

Sample correlation relies on products of deviations, so a few large joint deviations dominate the numerator and the variances, pushing the estimate toward the direction of the outliers, typically upward. Scale-free does not mean robust to outliers.

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