CAS 5A 11/98 Q.23 (question 4.85 in Mahler aggregate dist.

[Kenny]

I was told by one of the authors of one of the study guides to use the continuity correction for normal approximation, in the SOA exam setting, when it is stated in the problem. I did not verify that all solutions to previously published questions asking for normal approximation followed this guideline. If anyone has an example where it is not true, especially since the conversion, please let us know.

[shluffer]

The question in the first post of this thread uses the normal approximation of a discrete aggregate distibution (both the frq and sev are discrete) adn does not use the correction. This question is from before the conversion, but you have to remember that the CAS will not necesarily follow the same conventions that the SOA has. So for those of use taking the CAS exam, if there is a reason the CAS who originaly published this question did not use the correction we need to know it for the exam.

[Howard Mahler]

I have looked over the question, and now think I should have used the continuity correction. I will revise this for my Spring 2004 Study Aids.

The solution I provided was that provided at the time by the CAS to the 5A exam, from which the question was taken. I believe the reading was
Ch. 12 of Actuarial Mathematics, not on the current Course 3 Syllabus.

I am not sure why the CAS did not use the continuity correction, since it seems to be used in Ch. 12 of Actuarial Mathematics.

I would use the continuity correction if asked a similar question on Course 3, either CAS or SOA.

Howard Mahler