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
  #21  
Old 06-22-2018, 07:53 PM
Vorian Atreides's Avatar
Vorian Atreides Vorian Atreides is online now
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: 62,464
Default

Quote:
Originally Posted by Marcie View Post
Only the ones using red ink, Ms Marler.
__________________
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
  #22  
Old 06-22-2018, 07:59 PM
Marcie's Avatar
Marcie Marcie is offline
Member
CAS
 
Join Date: Feb 2015
Posts: 7,923
Default

Quote:
Originally Posted by Vorian Atreides View Post
In most cases, there really isn't one.

But that doesn't mean the DOI won't allow it for a "small company" to be competitive with large, national insurers.
They might even allow it for large, national insurers to be competitive with "small companies".
Reply With Quote
  #23  
Old 06-22-2018, 08:23 PM
whoanonstop's Avatar
whoanonstop whoanonstop is offline
Member
Non-Actuary
 
Join Date: Aug 2013
Location: Los Angeles, CA
Studying for Spark / Scala
College: College of William and Mary
Favorite beer: Orange Juice
Posts: 5,833
Blog Entries: 1
Default

Quote:
Originally Posted by Tacoactuary View Post
For those who have designed a binary retention model before, what kind of AUC (or any other measurement) should be considered respectable? I'm sure there will be plenty of variation by company and line of business, but any general targets would be helpful in determining if I'm heading in the right direction. Thanks
I'm going to guess that your outcomes are relatively balanced? Meaning that there is a good mix of 0's and 1's in your responses. If most people are being retained, say 95%+, then you've got other problems on your hand. But given your mentioned AUC, I doubt you're unbalanced.

Plus, if there is any plan to make business decisions based on results of this model, it is likely that your "errors" aren't equivalent and therefore you'll need to weight them accordingly. In that situation, AUC breaks down entirely.

Instead of creating hard class structures, 0 and 1, you could just look at probabilities of retention and use the a direct log loss in evaluating how good your model is doing.

Sadly, it is hard to compare AUC across multiple companies and multiple problems because they will vary heavily. If you want a better feel for what is "good" or not you might be better off just plotting a confusion matrix and seeing if the results seem acceptable. Comparing the AUCs and the confusion matrices for many different models might give you some better insight.

-Riley
__________________
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 03:05 PM.


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 0.14106 seconds with 11 queries