FRM Part I · FRM Exam Part I · Fundamentals of Probability
A bank's fraud model flags a transaction as suspicious. Historically, 2% of transactions are fraudulent. The model flags 90% of fraudulent transactions and also flags 5% of legitimate transactions. Given that a transaction is flagged, what is the probability that it is fraudulent (to the nearest 0.1%)?
The probability is about 26.9%. The total flag probability is 0.018 from fraud plus 0.049 from legitimate transactions, or 0.067. Dividing the fraud-and-flag probability of 0.018 by 0.067 gives 26.9%, because most flags come from the much larger legitimate group.
- A26.9%Correct
- B18.0%
- C90.0%
- D2.0%
Explanation
P(flag) = 0.02*0.90 + 0.98*0.05 = 0.018 + 0.049 = 0.067. P(fraud|flag) = 0.018/0.067 = 26.9%. The 18.0% option is the joint probability expressed as a percentage, which fails to divide by P(flag).
Did you get it right without looking?
One question tells you little. A timed set on Fundamentals of Probability shows your real accuracy, how long you take and where you lose marks.
More Fundamentals of Probability questions
- An insurer's claim count N has the following distribution: P(N=0)=0.5, P(N=1)=0.3, P(N=2)=0.2. What is Var(N)?
- A bank's fraud model flags a transaction as suspicious. Historically, 2% of transactions are fraudulent. The model flags 90% of fraudulent t…
- A trader states that events A and B are mutually exclusive, with P(A) = 0.20 and P(B) = 0.50. What is P(A | B), the probability of A given t…
- A risk analyst defines two events for next quarter: A = 'the portfolio loses more than 5%' and B = 'the portfolio gains more than 2%'. Which…
- A credit portfolio has 2% of loans in default. A screening model flags 90% of defaulted loans and also flags 8% of non-defaulted loans. Give…
- A credit officer partitions next year's outcomes for a corporate borrower into three mutually exclusive and exhaustive events: E1 = 'no chan…