Skip to content

FRM Part II · FRM Exam Part II · Credit Scoring and Retail Credit Risk Management

A bank's scorecard has two score bands. Band A has 1,000 accounts with a predicted PD of 2% and 30 observed defaults. Band B has 500 accounts with a predicted PD of 8% and 30 observed defaults. Which conclusion about calibration is best supported?

Band A defaults at 3.0% against a 2% prediction, so it is underpredicted, while Band B defaults at 6.0% against 8%, so it is overpredicted. Totals match at 60 defaults, but offsetting errors mask the problem. Ranking is still correct, so the issue is calibration.

  1. ABand A is underpredicted (observed 3.0% vs 2.0%) and Band B is overpredicted (observed 6.0% vs 8.0%), so calibration is imperfect even though the bands rank correctlyCorrect
  2. BBoth bands are well calibrated because total predicted defaults (60) equal total observed defaults (60)
  3. CBand A is overpredicted and Band B is underpredicted, so ranking is inverted
  4. DCalibration cannot be assessed without the AUC

Explanation

Band A: 30/1000 = 3.0% vs 2% predicted. Band B: 30/500 = 6.0% vs 8% predicted. Predicted total = 20 + 40 = 60, equal to observed 60, but offsetting errors hide band-level miscalibration. Ranking holds since 6% > 3%.

Did you get it right without looking?

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

Practise Credit Scoring and Retail Credit Risk ManagementFree diagnostic mock

More Credit Scoring and Retail Credit Risk Management questions