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

2017 ACTUARIAL SALARY SURVEYS
Contact DW Simpson for a Personalized Salary Survey

Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 12-01-2018, 02:32 PM
Colonel Smoothie's Avatar
Colonel Smoothie Colonel Smoothie is online now
Member
CAS
 
Join Date: Sep 2010
College: Jamba Juice University
Favorite beer: AO Amber Ale
Posts: 48,242
Default Support Vector Machines

Does anyone use SVM at work? I was reading a book and the author went over some examples where logistic regression would overfit, but SVM was more stable to outliers. That intrigues me. I have no experience with SVM at work, though.
__________________
Recommended Readings for the EL Actuary || Recommended Readings for the EB Actuary

Quote:
Originally Posted by Wigmeister General View Post
Don't you even think about sending me your resume. I'll turn it into an origami boulder and return it to you.
Reply With Quote
  #2  
Old 12-07-2018, 01:37 AM
Colonel Smoothie's Avatar
Colonel Smoothie Colonel Smoothie is online now
Member
CAS
 
Join Date: Sep 2010
College: Jamba Juice University
Favorite beer: AO Amber Ale
Posts: 48,242
Default

ok the book says, linear SVM is almost equivalent to L2-regularized logistic regression
__________________
Recommended Readings for the EL Actuary || Recommended Readings for the EB Actuary

Quote:
Originally Posted by Wigmeister General View Post
Don't you even think about sending me your resume. I'll turn it into an origami boulder and return it to you.
Reply With Quote
  #3  
Old 12-07-2018, 01:38 AM
Colonel Smoothie's Avatar
Colonel Smoothie Colonel Smoothie is online now
Member
CAS
 
Join Date: Sep 2010
College: Jamba Juice University
Favorite beer: AO Amber Ale
Posts: 48,242
Default

I guess the regularization is what reduces the overfitting
__________________
Recommended Readings for the EL Actuary || Recommended Readings for the EB Actuary

Quote:
Originally Posted by Wigmeister General View Post
Don't you even think about sending me your resume. I'll turn it into an origami boulder and return it to you.
Reply With Quote
  #4  
Old 12-09-2018, 08:03 PM
hostess's Avatar
hostess hostess is offline
Member
CAS
 
Join Date: Aug 2018
Posts: 396
Default

SVM seems similar to KNN imo
Reply With Quote
  #5  
Old 12-09-2018, 11:06 PM
AbedNadir's Avatar
AbedNadir AbedNadir is online now
Member
CAS SOA
 
Join Date: Mar 2014
Studying for enlightenment
Posts: 2,165
Default

haven't seen any real-life problems where it out-performs lightgbm. there may be some contrived examples where a kernel fits perfectly and makes separation but it would never happen on real data.
__________________
Reply With Quote
  #6  
Old 12-10-2018, 02:45 PM
Colonel Smoothie's Avatar
Colonel Smoothie Colonel Smoothie is online now
Member
CAS
 
Join Date: Sep 2010
College: Jamba Juice University
Favorite beer: AO Amber Ale
Posts: 48,242
Default

The book didn't really go into what the model form looks like. Are the parameters easy to explain? If not I suppose that is a good reason to stick with logistic regression.
__________________
Recommended Readings for the EL Actuary || Recommended Readings for the EB Actuary

Quote:
Originally Posted by Wigmeister General View Post
Don't you even think about sending me your resume. I'll turn it into an origami boulder and return it to you.
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 09:45 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 0.29407 seconds with 9 queries