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 our jobs meet your skills and objectives.


Reply
 
Thread Tools Search this Thread Display Modes
  #31  
Old 04-18-2019, 08:22 PM
MoralHazard MoralHazard is offline
Member
CAS
 
Join Date: Jul 2011
Favorite beer: Sam Adams Rebel Rouser
Posts: 110
Default

Quote:
Originally Posted by itGetsBetter View Post
Policy year structure is best because you have a perfect match between policy characteristics and the losses that resulted from them. An additional tip is to add policy year as a categorical variable in the model to control for trend without having to perform trend analysis.
With the C/AY structure you also have the policy characteristics matching the losses.

An advantage of C/AY is that if you are going to incorporate any trend or loss development, that tends to fit more naturally with accident years than policy years. Also makes it easier to reconcile the loss and exposure data in the model to financial statements/reports.
Reply With Quote
  #32  
Old 04-19-2019, 01:50 AM
AbedNadir's Avatar
AbedNadir AbedNadir is offline
Member
CAS SOA
 
Join Date: Mar 2014
Studying for FCAS
Posts: 2,584
Default

Quote:
Originally Posted by ALivelySedative View Post
No no, I mean basic stuff. Driving experience is not recorded as a raw value. The combination of drivers' experience on a policy is binned into some categorical value, which is then recorded for all risks on the policy. So not only is it not recorded for an individual, they're certainly not matched to an actual exposure (car-year). I'm just trying to convince people that this is a bad approach to our own data.
so if someone wants to add a vehicle to their policy, how do you rate for that?
__________________
Reply With Quote
  #33  
Old 04-25-2019, 09:06 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: 65,751
Default

Resource I came across for the CSPA Exam 3 material. Contains R code for doing the following:
  1. Gini coefficient
  2. (Plotting) Lorenz Curve
  3. Lift charts (including double lift charts)
  4. Receiver Operator Characteristic (ROC) curve

https://thecasinstitute.org/wp-conte...n-01162019.pdf
__________________
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
  #34  
Old 04-26-2019, 10:59 AM
Actuarially Me Actuarially Me is offline
Member
CAS
 
Join Date: Jun 2013
Posts: 191
Default

So after underwriters make their adjustments to the factors, how can you easily compare the original model with the new model?

I can do it manually in Excel by exporting the most recent credible policy year, creating a "code" column that contains the variable name concatenated with the variable value, using a Vlookup to pull in coefficients, and then manually multiplying the factors together to get a prediction. Then from there, can calculate gini and lift charts.

I can replicate that process in R too with some more work, but is there any easier way I'm overlooking?
Reply With Quote
  #35  
Old 04-26-2019, 12:45 PM
examsarehard examsarehard is offline
Member
CAS
 
Join Date: May 2011
Posts: 597
Default

In a GLM, you specify the underwriter adjustments as offsets.
Reply With Quote
  #36  
Old 04-27-2019, 07:58 PM
jerrytuttle jerrytuttle is offline
Member
CAS
 
Join Date: Oct 2001
Posts: 290
Default

Hi. Is there a publicly available insurance database, possibly with fictional data, to play with predictive modeling?
__________________
Thanks,
Jerry
Reply With Quote
  #37  
Old 04-29-2019, 09:06 PM
Avi's Avatar
Avi Avi is offline
Wiki Contributor
Site Supporter
Site Supporter
CAS AAA
 
Join Date: Aug 2002
Location: NY
Studying for the rest of my life.
College: Alumnus - Queens College - CUNY
Favorite beer: Stone Ruination IPA
Posts: 14,212
Blog Entries: 3
Default

data.table >> dplyr so long as the data fits in RAM.
__________________
All scientists defer only to physicists
Physicists defer only to mathematicians
Mathematicians defer only to G-d!

--with apologies to Dr. Leon Lederman
Reply With Quote
  #38  
Old 04-30-2019, 11:14 AM
Actuarially Me Actuarially Me is offline
Member
CAS
 
Join Date: Jun 2013
Posts: 191
Default

Quote:
Originally Posted by Avi View Post
data.table >> dplyr so long as the data fits in RAM.
True, but not everyone knows data.table well enough to tech review something. I'd rather have longer, readable code that more people can understand. Dplyr is similar enough to SQL
Reply With Quote
  #39  
Old 04-30-2019, 11:16 AM
Actuarially Me Actuarially Me is offline
Member
CAS
 
Join Date: Jun 2013
Posts: 191
Default

Quote:
Originally Posted by jerrytuttle View Post
Hi. Is there a publicly available insurance database, possibly with fictional data, to play with predictive modeling?
insuranceData package in R has data sets.

There are some insurance data sets in Kaggle, but they usually have generic named fields.
Reply With Quote
  #40  
Old 05-01-2019, 12:44 AM
Avi's Avatar
Avi Avi is offline
Wiki Contributor
Site Supporter
Site Supporter
CAS AAA
 
Join Date: Aug 2002
Location: NY
Studying for the rest of my life.
College: Alumnus - Queens College - CUNY
Favorite beer: Stone Ruination IPA
Posts: 14,212
Blog Entries: 3
Default

Quote:
Originally Posted by Actuarially Me View Post
True, but not everyone knows data.table well enough to tech review something. I'd rather have longer, readable code that more people can understand. Dplyr is similar enough to SQL
Personally, I find the data.table syntax easier to understand than the dplyr set of verbs, but that may be familiarity. It certainly is more compact, and can "do more with less"
__________________
All scientists defer only to physicists
Physicists defer only to mathematicians
Mathematicians defer only to G-d!

--with apologies to Dr. Leon Lederman
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 09:14 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 0.42532 seconds with 9 queries