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

DW Simpson International Actuarial Jobs
Canada  Asia  Australia  Bermuda  Latin America  Europe


Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 10-30-2019, 05:20 PM
AbedNadir's Avatar
AbedNadir AbedNadir is offline
Member
CAS SOA
 
Join Date: Mar 2014
Studying for FCAS
Posts: 2,714
Default Gini index interpretation

So you have an auto book of business worth X amount of money and you have two models, one with a gini index of .42 and one with .44. People looking at those numbers think, we spent Y months working on this models for a measly .02?!?!

Is anyone here aware of a way to interpret this in terms of dollars? For example, could we say something like, .02 is equal to saving Z dollars in loss dollars a year for every P amount of premium?
__________________
Reply With Quote
  #2  
Old 10-31-2019, 10:20 AM
Heywood J Heywood J is offline
Member
CAS
 
Join Date: Jun 2006
Posts: 4,101
Default

Quote:
Originally Posted by AbedNadir View Post
So you have an auto book of business worth X amount of money and you have two models, one with a gini index of .42 and one with .44. People looking at those numbers think, we spent Y months working on this models for a measly .02?!?!

Is anyone here aware of a way to interpret this in terms of dollars? For example, could we say something like, .02 is equal to saving Z dollars in loss dollars a year for every P amount of premium?
Not without making a dozen flimsy assumptions. Probably the best thing to do is to show a double lift chart, where you can show how substantial the differences in predictions are, and they will be substantial for 0.44 vs 0.42 difference, assuming that's calculated on a large enough dataset.
Reply With Quote
  #3  
Old 10-31-2019, 10:30 AM
Vorian Atreides's Avatar
Vorian Atreides Vorian Atreides is offline
Wiki/Note Contributor
CAS
 
Join Date: Apr 2005
Location: As far as 3 cups of sugar will take you
Studying for ACAS
College: Hard Knocks
Favorite beer: Most German dark lagers
Posts: 66,879
Default

First, what other measures of "performance" was looked at?

Gini Index is a good for assessing if one model is able to better differentiate "high" vs. "low" risks (assuming that your model is predicting insured losses).

But I don't think you're going to find a good (direct) connection between the difference in the Gini Index and some monetary result since a lot depends on being able to get the needed rate for the higher risks and still be able to sell (i.e., get more) of the lower risk; and this will require looking at other mechanisms than the Gini Index.
__________________
I find your lack of faith disturbing

Why should I worry about dying? Itís not going to happen in my lifetime!


Freedom of speech is not a license to discourtesy

#BLACKMATTERLIVES
Reply With Quote
  #4  
Old 10-31-2019, 09:04 PM
Glenn Meyers Glenn Meyers is offline
Member
 
Join Date: Mar 2002
Posts: 277
Default

If you are interested in the dollar effect of lift, you should at the Lorenz curve underlying the Gini index. Pick a point on the horizontal (% Premium) axis. The value on the vertical (% loss) will give the % of loss associated with the % of premium. If the % of loss is less than the % of premium, you should consider using the refined class plan. Consideration can include adverse selection, price elasticity and the cost of implementing the refined class plan.

You may want to see my Actuarial Review article on the “Value of Lift”

https://www.casact.org/newsletter/in...iewart&id=5584

This was written before our work on the Gini Index. We move on to the Gini Index because we (ISO) did not want to get hung up on the above considerations. That being said, I think it is appropriate for companies to make those considerations.
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 07:41 AM.


Powered by vBulletin®
Copyright ©2000 - 2020, 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 0.15935 seconds with 11 queries