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


Upload your resume securely at https://www.dwsimpson.com
to be contacted when new jobs meet your skills and objectives.


Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 10-03-2018, 10:38 AM
Bingboing's Avatar
Bingboing Bingboing is offline
Member
SOA
 
Join Date: Nov 2013
Posts: 7,775
Default Reserving methods that don't rely on Paid patterns

We are about to have a shift in our historical payment patterns and are looking into alternative methods for reserving that don't rely (fully) on the paid data we are seeing. Note this pattern change is going to be abrupt and continually changing over a few years worth of time...

So mostly it seems like I am looking for some sort of projection method.
Loss Ratio
Trending Incurreds
Regression on historical incurreds
??
other ideas?

I'm looking for more methods to add to our models to give us a good idea of what's going on. There are going to be issues with all of the methods I think, but understanding what's out there and how to adjust is what I am looking for.

Thanks.
Reply With Quote
  #2  
Old 10-03-2018, 10:39 AM
Bingboing's Avatar
Bingboing Bingboing is offline
Member
SOA
 
Join Date: Nov 2013
Posts: 7,775
Default

Quote:
Originally Posted by Bingboing View Post
We are about to have a shift in our historical payment patterns and are looking into alternative methods for reserving that don't rely (fully) on the paid data we are seeing. Note this pattern change is going to be abrupt and continually changing over a few years worth of time...

So mostly it seems like I am looking for some sort of projection method.
Loss Ratio
Trending Incurreds
Regression on historical incurreds
??
other ideas?

I'm looking for more methods to add to our models to give us a good idea of what's going on. There are going to be issues with all of the methods I think, but understanding what's out there and how to adjust is what I am looking for.

Thanks.
I've cross posted this in both Health and P&C since they are similar.
Reply With Quote
  #3  
Old 10-03-2018, 10:46 AM
Marcie's Avatar
Marcie Marcie is offline
Member
CAS
 
Join Date: Feb 2015
Posts: 8,859
Default

What about a Berquist-Sherman type adjustment of past data to the latest (or future) patterns?
Reply With Quote
  #4  
Old 10-03-2018, 11:29 AM
Westley Westley is offline
Member
 
Join Date: Nov 2001
Posts: 27,664
Default

For the incurred methods you list, I assume you mean
incurred = paid + case
(as opposed to an accounting terminology, where incurred = ultimate)
So, they'll still be distorted by the paid changes, right? Just making sure I understand.

How late do your claims report? Is it worth doing separate methodologies based on pure IBNR separate from development on known? Development on know could use either a trended severity approach, or if you expect case reserving adequacy to be unchanged, could do something there. And pure IBNR shouldn't be distorted?
Reply With Quote
  #5  
Old 10-03-2018, 11:55 AM
Abelian Grape's Avatar
Abelian Grape Abelian Grape is offline
Meme-ber                         Meme-ber
CAS
 
Join Date: Jul 2014
Favorite beer: Allagash Curieux
Posts: 42,036
Default

Quote:
Originally Posted by Marcie View Post
What about a Berquist-Sherman type adjustment of past data to the latest (or future) patterns?
Is this actually usable in practice?
__________________
Quote:
Originally Posted by Pension.Mathematics View Post
what's your problem man?
Reply With Quote
  #6  
Old 10-03-2018, 11:58 AM
Westley Westley is offline
Member
 
Join Date: Nov 2001
Posts: 27,664
Default

Are you asking about Berquist-Sherman in general? Not sure it works for the OP's problem but have used it before. If you can document change in claim closure rate, and a few other things hold up (few or no partial payments, no change in claims priority, etc), I think it's a good adjustment for closure speed.

I've tried to use the case reserve adequacy methods from that paper, and IIRC have never had them produce useable results.
Reply With Quote
  #7  
Old 10-03-2018, 12:33 PM
PeppermintPatty's Avatar
PeppermintPatty PeppermintPatty is offline
Member
CAS
 
Join Date: Sep 2001
Posts: 39,126
Default

I've tried Berquist Sherman adjustments a few times. I did not have any luck getting more credible answers adjusting either closing patterns or case reserve adequacy in a long-tailed GL line, but I tried the disposal adjustment on a personal auto liability book once, and it worked beautifully, and it looked like that explained about 70%-90% of the variation in age-to-age development.

I see the OP is SoA, and probably in health, so I'd like to point them to the CAS exam 5 reserving syllabus
https://www.casact.org/admissions/sy...x.cfm?fa=exam5

bingboing, the Friedland paper is a free download, and even if you don't use any of the methods in it exactly, it should give you some ideas.
Reply With Quote
  #8  
Old 10-03-2018, 02:34 PM
Bingboing's Avatar
Bingboing Bingboing is offline
Member
SOA
 
Join Date: Nov 2013
Posts: 7,775
Default

Quote:
Originally Posted by Westley View Post
For the incurred methods you list, I assume you mean
incurred = paid + case
(as opposed to an accounting terminology, where incurred = ultimate)
So, they'll still be distorted by the paid changes, right? Just making sure I understand.

How late do your claims report? Is it worth doing separate methodologies based on pure IBNR separate from development on known? Development on know could use either a trended severity approach, or if you expect case reserving adequacy to be unchanged, could do something there. And pure IBNR shouldn't be distorted?
Yep, talking about our estimation of the incurred based on the current paid.
It's health claims, so we are pretty complete relatively quickly (98% complete in 6 months or so), but we will still be receiving some claims and recoveries for years (we only look at about 3 years) beyond the incurred dates.
Reply With Quote
  #9  
Old 10-03-2018, 02:37 PM
Bingboing's Avatar
Bingboing Bingboing is offline
Member
SOA
 
Join Date: Nov 2013
Posts: 7,775
Default

Quote:
Originally Posted by PeppermintPatty View Post
I've tried Berquist Sherman adjustments a few times. I did not have any luck getting more credible answers adjusting either closing patterns or case reserve adequacy in a long-tailed GL line, but I tried the disposal adjustment on a personal auto liability book once, and it worked beautifully, and it looked like that explained about 70%-90% of the variation in age-to-age development.

I see the OP is SoA, and probably in health, so I'd like to point them to the CAS exam 5 reserving syllabus
https://www.casact.org/admissions/sy...x.cfm?fa=exam5

bingboing, the Friedland paper is a free download, and even if you don't use any of the methods in it exactly, it should give you some ideas.
Yeah, I think I looked through some of these, but was hoping to get more experience based answers (and quicker) than looking through all of those texts. If you have a specific one in mind, I don't mind reading through, but not sure I have time to read all of them
Reply With Quote
  #10  
Old 10-03-2018, 02:37 PM
Bingboing's Avatar
Bingboing Bingboing is offline
Member
SOA
 
Join Date: Nov 2013
Posts: 7,775
Default

Sorry, missed that last sentence, I'll take a look there.
Thanks.
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 12:40 AM.


Powered by vBulletin®
Copyright ©2000 - 2018, 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 0.24993 seconds with 10 queries