Fall 2016 Q27b)

judyohjudyoh
in F-15-Evaluation
Screen Shot 2023-08-04 at 8.37.29 AM.png
Screen Shot 2023-08-04 at 8.37.44 AM.png

I am not understanding the solution. In general, I can understand the limitations but I do not understand the alternative calculations in sample answer 2-4.

Sample answer 2: is the solution suggesting you project closed claim counts to ultimate and compare those ultimates to the selected ultimates? And instead of using reported claim count cdfs, use the closed claim count cdfs, replacing the selected ultimates with paid claim count ultimates?

Sample answer 3: isn't a change in number of small claims similar to sample answer 5? I thought Berq Sherman method adjusts for changes in settlement rate, while for changes in settlement of small/large claims you should separate data into small/large claims and apply BS adjustment to each group

Sample answer 4: what is an incremental closed method? Using CDR to estimate incremental closed counts as we do in FS method?

Apologies for the long question, overall I would like further clarification.

Comments

  • GrahamGraham
    edited August 2023

    For sample answer 2:

    • I think they meant to say use the "reported claim count triangle" not the paid (closed) claim count triangle. Part (a) asks about reported emergence not paid emergence.

    For sample answer 3:

    • If the problem is that there are more small claims then a better answer is to separate the count data by claim size. You would have to analyze 2 (or more triangles) but the data would be more homogenous. (I think the candidate who wrote that answer was just grasping for a solution and made something up that doesn't necessarily make sense.)

    For sample answer 4:

    For sample answer 5:

    • This does seem similar to the limitation stated in sample answer 3 - that there has been some sort of a change in the mix of business.

    Final Comment on this exam question:

    • I think sample answer 1 is the best because it's the simplest to understand and to explain.
    • The "limitations" in the other sample answers are all correct, but it isn't as easy to explain the alternative methods. I think that's the reason the graders seemed to be lenient. They realized there was no simple alternative as there was for sample answer 1 so as long as you suggested something that was at least plausible, they gave you credit.
    • Important note: Don't get too bogged down with all these sample answers. Just pick 1 or 2 that make the most sense to you, study those, then don't worry too much about the others. While it's often helpful to study various different solutions to a problem, it's usually the case that just 1 (or maybe 2) are better than the others, and that the others are not really worth spending time on. These sample answers are often just copied from someone's actual exam paper so they aren't always very good answers. They are just for illustration, and they weren't necessarily even given full credit.
  • judyohjudyoh

    Thank you for taking the time to respond and the tips at the end, I will keep them in mind :)

  • GrahamGraham

    You're welcome! 🙂

Sign In or Register to comment.