Actuarial Outpost Aggregating Trend Experience
 Register Blogs Wiki FAQ Calendar Search Today's Posts Mark Forums Read
 FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

 General Actuarial Non-Specific Actuarial Topics - Before posting a thread, please browse over our other sections to see if there is a better fit, such as Careers - Employment, Actuarial Science Universities Forum or any of our other 100+ forums.

#1
08-14-2018, 11:07 AM
 UncleBocky Member SOA Join Date: May 2009 Posts: 109
Aggregating Trend Experience

See attached excel file. I have two plans, each with different factors that predict the claim cost. When aggregating claims experience and breaking down the trend into different factors at an aggregate level, what's the best way to do this? Member-weighted? Claims-weighted? In the file I have a couple examples of ways to aggregate it if using a member-weighted approach, but unsure which method is better or if a claims-weighted approach should be used. Any help is appreciated.
Attached Files
 Aggregating Trends Example.xlsx (18.1 KB, 31 views)
#2
08-14-2018, 02:04 PM
 packman Member SOA AAA Join Date: Nov 2007 Posts: 231

I think your option 2 is correct.
#3
08-15-2018, 06:30 PM
 connordavoice SOA Join Date: Jul 2018 College: Abilene Christian University, Alumnus Favorite beer: Does Rum Count? Posts: 6

So that I understand what's going on, I have a few questions.
I believe I understand that the Claims column refers to the number of claims in a given month, provided a certain Norm and a pair of factors. So what does the Norm refer to, expected number of claims? Are the values for the factors arbitrarily chosen in this example or is there a reason to them?
Aggregate seems very similar to an average, but weighted by number of members it appears. What perplexes me is the "Other" column in the second aggregate option. Is this another factor derived from the other material in option 2?
#4
08-15-2018, 06:59 PM
 DoctorNo Member SOA AAA Join Date: Dec 2004 Location: Denver, CO College: Western Washington, Colorado-Boulder Posts: 13,472

The question I usually ask when doing this is: how is the result going to be used? If it's going to be used to influence claim costs, then there's your answer.
__________________
The opinions of Doctor No do not necessarily represent the opinions of mathematicians or consulting actuaries. Facts cited by Doctor No are not necessarily facts. Find me on Twitter: @NorrisDoug. If you send me a LinkedIn invitation, please let me know who you are (unless it's obvious how I know you).