FRM Part II · FRM Exam Part II · Credit Scoring and Retail Credit Risk Management
A bank's scorecard has an area under the ROC curve (AUC) of 0.82 on its validation sample. What is the corresponding Gini coefficient (accuracy ratio)?
The Gini coefficient equals 2 × AUC − 1, so with an AUC of 0.82 it is 0.64. Doubling the excess of AUC over the random-model value of 0.5 rescales discriminatory power so that a random model scores zero and a perfect one scores one.
- A0.32
- B0.64Correct
- C0.82
- D0.18
Explanation
Gini = 2 × AUC − 1 = 2 × 0.82 − 1 = 0.64. Option A subtracts 0.5 without doubling. Option C confuses Gini with AUC. Option D is 1 − AUC.
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.
More Credit Scoring and Retail Credit Risk Management questions
- A retail lender's credit scorecard excludes race, gender and religion as inputs. Regulators nonetheless find that approval rates for one pro…
- A bank tests a scorecard for fairness. Group A has 1,000 applicants with 600 approved. Group B has 500 applicants with 200 approved. Using t…
- In a scorecard development sample, a bin of an applicant characteristic contains 200 goods and 20 bads. The whole sample contains 4,000 good…
- A card issuer finds that its scorecard, built three years ago, still ranks customers well (stable Gini) but the observed default rate in the…
- A card issuer notices that customers who recently had their credit lines increased through an automated program show a higher subsequent def…
- A lender rejects a loan applicant using an automated scoring model. Under typical adverse action and consumer protection requirements, which…