IAI Actuarial Core Principles · Risk Modelling and Survival Analysis · Elementary principles of machine learning
A binary classifier for predicting policy lapse is tested on 200 policies. It gives 40 true positives, 20 false positives, 10 false negatives and 130 true negatives. What is the precision of the classifier?
Precision is true positives divided by all predicted positives, which is 40/(40+20) = 0.667. The figure 0.80 is the recall, and 0.85 is the overall accuracy, so those options use the wrong measure.
- A40/50 = 0.80
- B40/60 = 0.667Correct
- C170/200 = 0.85
- D130/150 = 0.867
- 40/200 = 0.20
Explanation
Precision = TP/(TP+FP) = 40/(40+20) = 40/60 = 0.667. The value 0.80 is recall, TP/(TP+FN) = 40/50. The value 0.85 is accuracy, (40+130)/200.
Did you get it right without looking?
One question tells you little. A timed set on Elementary principles of machine learning shows your real accuracy, how long you take and where you lose marks.
More Elementary principles of machine learning questions
- When building a predictive model for insurance data, which action best reduces the risk of data leakage?
- A health insurer groups 50,000 customers into segments using only age, income and claim frequency, with no predefined segment labels. Which …
- A pricing team fits a claim-severity model and finds that its error on the data used for fitting is very small, but its error on a separate …
- A modeller fits polynomial regressions of increasing degree to the same training data. The training error falls steadily with degree, but th…
- In ridge regression, a penalty equal to a positive constant λ times the sum of squared coefficients is added to the residual sum of squares.…
- A model predicting whether a policy lapses is tested on 1,000 policies, of which 100 actually lapse. The model predicts lapse for 80 policie…