Determine which of the following statements regarding statistical learning methods is/are true.

• Methods that are highly interpretable are more likely to be highly flexible.
• When inference is the goal, there are clear advantages to using a lasso method versus a bagging method.
• Using a more flexible method will produce a more accurate prediction against unseen data.

• I only
• II only
• III only
• I, II and III
• The correct answer is not given by (A), (B), (C), or (D).

Copyright 2023. The Society of Actuaries, Schaumburg, Illinois. Reproduced with permission.

Key: B

I is false. Highly flexible models are harder to interpret. For example, a ninth degree polynomial is harder to interpret than a straight line.

II is true. Inference is easier when using simple and relatively inflexible methods. Lasso is simpler and less flexible than bagging.

III is false. Flexible methods tend to overfit the training set and be less accurate when applied to unseen data.

Copyright 2023. The Society of Actuaries, Schaumburg, Illinois. Reproduced with permission.