FRM Part II · FRM Exam Part II · Advances in Artificial Intelligence: Implications for Capital Markets Activities
A surveillance team tests an AI model on 10,000 trading days, of which 200 contain true market abuse. The model flags 250 days, and 150 of these are true abuse. What are the model's precision and recall, respectively?
Precision is 60% because 150 of the 250 flagged days were genuine abuse. Recall is 75% because the model caught 150 of the 200 actual abuse days. Reversing the denominators would wrongly give 75% and 60%.
- A60% and 75%Correct
- B75% and 60%
- C60% and 1.5%
- D75% and 1.5%
Explanation
Precision = true positives / flagged = 150/250 = 60%. Recall = true positives / actual abuse = 150/200 = 75%. Swapping the two gives 75% and 60%, which is the common mistake; 1.5% divides by all 10,000 days.
Did you get it right without looking?
One question tells you little. A timed set on Advances in Artificial Intelligence: Implications for Capital Markets Activities shows your real accuracy, how long you take and where you lose marks.
More Advances in Artificial Intelligence: Implications for Capital Markets Activities questions
- A broker-dealer replaces its keyword-based trade communications surveillance with a natural language processing model that flags context-dep…
- A trading firm uses a deep neural network to generate short-term alpha signals. Risk management cannot explain why the model takes large pos…
- A risk manager notes that AI models used for credit and market decisions may perform poorly in a regime unlike the data they were trained on…
- A trading firm uses a natural language processing model to turn news and social media into trading signals. Which control is most appropriat…
- A bank trains a machine-learning credit-trading model on ten years of market data in which volatility was persistently low. During a sudden …
- A trading desk uses a large language model to summarize central bank communications and generate sentiment scores for use in a systematic st…