Skip to content

FRM Part II · FRM Exam Part II · Credit Risk

In developing a retail credit scorecard using logistic regression, an analyst finds that the reject-inference step is needed. What is the main reason?

Reject inference is needed because default outcomes are observed only for applicants the bank accepted, so a model built on that sample is biased toward the past acceptance policy. Inferring how rejected applicants would have performed reduces this selection bias and improves the scorecard's applicability to all applicants.

  1. APerformance data exist only for accepted applicants, which biases the sampleCorrect
  2. BLogistic regression cannot handle binary outcomes
  3. CRejected applicants always default at higher rates than accepted ones
  4. DScore cut-offs must be set at the mean score

Explanation

Outcomes are observed only for applicants who were granted credit, so the development sample suffers selection bias. Reject inference attempts to estimate how rejected applicants would have performed. Logistic regression is designed for binary outcomes.

Did you get it right without looking?

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

Practise Credit RiskFree diagnostic mock

More Credit Risk questions