An analyst is modeling the probability of a certain phenomenon occurring. The analyst has observed that the simple linear model currently in use results in predicted values less than zero and greater than one.

Determine which of the following is the most appropriate way to address this issue.

• Limit the data to observations that are expected to result in predicted values between 0 and 1.
• Consider predicted values below 0 as 0 and values above 1 as 1.
• Use a logit function to transform the linear model into only predicting values between 0 and 1.
• Use the canonical link function for the Poisson distribution to transform the linear model into only predicting values between 0 and 1.
• None of the above.

Key: C

(A) is not appropriate because removing data will likely bias the model estimates.

(B) is not appropriate because altering data will likely bias the model estimates.

(C) is correct.

(D) is not appropriate because the canonical link function is the logarithm, which will not restrict values to the range zero to one.

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