Skip to content

FRM Part II · FRM Exam Part II · Credit Scoring and Rating

A bank validates its corporate rating model by plotting the cumulative proportion of all borrowers (ranked from worst to best score) against the cumulative proportion of defaulters captured. The resulting curve lies well above the diagonal line. What does this indicate?

A CAP curve lying above the diagonal shows the model has discriminatory power better than random, because defaulters are concentrated among the worst-scored borrowers. It measures ranking ability, not calibration of PD levels, so it cannot show predicted PDs match observed default rates.

  1. AThe model has discriminatory power better than random assignment of scoresCorrect
  2. BThe model is well calibrated because predicted PDs equal observed default rates
  3. CThe rating system has no migration over time
  4. DThe model is overfitted to the development sample

Explanation

A cumulative accuracy profile (CAP) curve above the diagonal means defaulters are concentrated among the worst-scored borrowers, so the model ranks risk better than random. It says nothing about whether PD levels are calibrated, which requires comparing predicted and realized default rates. Overfitting and migration cannot be inferred from the curve alone.

Did you get it right without looking?

One question tells you little. A timed set on Credit Scoring and Rating shows your real accuracy, how long you take and where you lose marks.

More Credit Scoring and Rating questions