Skip to content

FRM Part II · FRM Exam Part II · Case Study: Financial Crime and Fraud

A bank's fraud analytics team reviews a transaction-monitoring model. At a given alert threshold the model generated 1,000 alerts, of which 50 were confirmed as fraud. Investigators later learned that 150 frauds occurred in the period in total, so 100 were not alerted. What are the alert precision and the detection rate (recall)?

Precision is 5% and recall is 33.3%. Precision is 50 confirmed frauds divided by 1,000 alerts, while recall is 50 detected frauds divided by the 150 total frauds. Most alerts are false positives, and two-thirds of frauds went undetected.

  1. APrecision 5%; recall 33.3%Correct
  2. BPrecision 33.3%; recall 5%
  3. CPrecision 5%; recall 50%
  4. DPrecision 15%; recall 33.3%

Explanation

Precision = confirmed frauds among alerts / alerts = 50/1,000 = 5%. Recall = frauds detected / total frauds = 50/150 = 33.3%. Check: missed 100 plus detected 50 equals 150. The 15% option wrongly uses 150/1,000, dividing total frauds by alerts, and the 50% option divides detected by missed frauds.

Did you get it right without looking?

One question tells you little. A timed set on Case Study: Financial Crime and Fraud shows your real accuracy, how long you take and where you lose marks.

Practise Case Study: Financial Crime and FraudFree diagnostic mock

More Case Study: Financial Crime and Fraud questions