Actuarial Outpost
 
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Property - Casualty / General Insurance
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

ACTUARIAL SALARY SURVEYS
Contact DW Simpson for a Personalized Salary Survey

Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 06-19-2014, 03:40 PM
Final Boss's Avatar
Final Boss Final Boss is offline
Member
CAS
 
Join Date: Nov 2012
Posts: 295
Default Allocating Development to Indv Losses for Segmentation

I'm getting a dataset ready for segmentation analysis. I have incurred losses by policy, however, need to develop them to ultimate. I was curious how some of you go about developing individual losses. If I roll them up, develop the group using a traditional chain-ladder method, how would you recommend I re-allocate the group's development back down to the individual policy level? Not really sure what to search for in the literature. Thanks in advance.
Reply With Quote
  #2  
Old 06-19-2014, 04:10 PM
MountainHawk's Avatar
MountainHawk MountainHawk is offline
Member
CAS AAA
 
Join Date: Dec 2001
Location: Salem, MA
Studying for Nothing!!!!
College: Lehigh University Alum
Favorite beer: Yuengling
Posts: 64,850
Default

It's really hard. Do you have pure IBNR? If not, then you can just allocate IBNR by case outstanding, subject to any limits and it's not awful. If you do, then you have to allocate the pure IBNR to the non-existent claims, and then allocate the rest of the IBNR by case outstanding.
__________________

Play in the AO Prediction Game now!



1
Reply With Quote
  #3  
Old 06-19-2014, 04:44 PM
Arthur Itas's Avatar
Arthur Itas Arthur Itas is offline
Member
CAS
 
Join Date: Feb 2004
Location: Atlanta
Studying for Prostate exam
College: Hard Knocks
Posts: 20,630
Default

I don't like the concept of policy level IBNR, but many carriers seem to want to do it. I understand why it appears desirable, as you can then roll up results anyway you like: broker, branch, segment, underwriter,...

Aggregate development does not translate well to individual account development. MH hits on one important concept: pure IBNR. Other questions to consider: do you want to cap losses at policy limit as part of the project? Develop individual closed claims if you are using the chain ladder incurred method? Allow one potentially fortuitous claim to drive results for a given segment?
Reply With Quote
  #4  
Old 06-19-2014, 05:19 PM
Final Boss's Avatar
Final Boss Final Boss is offline
Member
CAS
 
Join Date: Nov 2012
Posts: 295
Default

Quote:
Originally Posted by Arthur Itas View Post
I don't like the concept of policy level IBNR, but many carriers seem to want to do it. I understand why it appears desirable, as you can then roll up results anyway you like: broker, branch, segment, underwriter,...

Aggregate development does not translate well to individual account development. MH hits on one important concept: pure IBNR. Other questions to consider: do you want to cap losses at policy limit as part of the project? Develop individual closed claims if you are using the chain ladder incurred method? Allow one potentially fortuitous claim to drive results for a given segment?
I can already see the group develops upwards, and it seems kind of silly to me to expect unfavorable development on limit losses, so I definitely want to adjust for limits. Specifically, I was thinking to remove features with a case reserve >= Limit - Paid-to-Date; then develop those under [limit - PTD], then add back the case incurred for those at [limit - PTD].

I'm more worried about the pure IBNR. I've got too many rating variables to say something like my class A exposures need $.02 for pure IBNR at 18 months, my class B exposures need $.50 for IBNR, etc.
Reply With Quote
  #5  
Old 06-19-2014, 05:33 PM
Colymbosathon ecplecticos's Avatar
Colymbosathon ecplecticos Colymbosathon ecplecticos is offline
Member
 
Join Date: Dec 2003
Posts: 6,133
Default

Quote:
Originally Posted by Final Boss View Post
I can already see the group develops upwards, and it seems kind of silly to me to expect unfavorable development on limit losses, so I definitely want to adjust for limits. Specifically, I was thinking to remove features with a case reserve >= Limit - Paid-to-Date; then develop those under [limit - PTD], then add back the case incurred for those at [limit - PTD].
Don't forget that the dollars of development that you don't allow on limit losses need to be made up on the other losses.

For example:

Case = 1100 (1000 + 100) 1000 is limit

Dev factor = 1.2

Ult = 1320

Since you aren't letting the 1000 grow, your dev factor for non-limit losses needs to be 3.2!
__________________
"What do you mean I don't have the prerequisites for this class? I've failed it twice before!"


"I think that probably clarifies things pretty good by itself."

"I understand health care now especially very well."
Reply With Quote
  #6  
Old 06-19-2014, 07:01 PM
ActuaryMAD ActuaryMAD is offline
Member
CAS
 
Join Date: Dec 2011
Posts: 138
Default

At my current employer, the allocation of IBNR is done as a ratio to earned premium. There are 3 business segments which have different IBNR values as % of premium.

IMO, it's not an optimal way to allocate IBNR - but with that being said, no major decisions are made based on that allocation. It's just a way to determine expected ultimate loss ratio for less than the book in total.
Reply With Quote
  #7  
Old 06-27-2014, 05:38 AM
Sherwin's Avatar
Sherwin Sherwin is offline
Member
CAS SOA
 
Join Date: Jun 2006
Location: China Reinsurer
Posts: 220
Blog Entries: 10
Default

There is a paper called "Reserving on a contract-by-contract basis" on the website of UK institute. You can search it.
__________________
To Make the Casualty Actuarial Profession More Internationalized.
Reply With Quote
  #8  
Old 11-24-2014, 04:12 PM
Final Boss's Avatar
Final Boss Final Boss is offline
Member
CAS
 
Join Date: Nov 2012
Posts: 295
Default

Quote:
Originally Posted by ActuaryMAD View Post
At my current employer, the allocation of IBNR is done as a ratio to earned premium. There are 3 business segments which have different IBNR values as % of premium.

IMO, it's not an optimal way to allocate IBNR - but with that being said, no major decisions are made based on that allocation. It's just a way to determine expected ultimate loss ratio for less than the book in total.
Sorry to re-hash an old thread, but I was playing around with the quoted idea and ran into an issue. If you allocate pure IBNR based on this method what would the loss cost distribution look like? Instead of say 90% of exposures having 0 loss and 10% having some loss (i.e. Tweedie) 100% of exposures in recent AY's have some loss. Additionally, I run into a similar problem when I try to split loss cost into its components: frequency and severity. Any thoughts?

So do most of you just stop your dataset say 18 months prior to the evaulation date and assume no pure IBNR and just develop case on a report period basis?
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


All times are GMT -4. The time now is 03:50 PM.


Powered by vBulletin®
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
*PLEASE NOTE: Posts are not checked for accuracy, and do not
represent the views of the Actuarial Outpost or its sponsors.
Page generated in 0.18855 seconds with 11 queries