Actuarial Outpost
 
Go Back   Actuarial Outpost > Exams - Please Limit Discussion to Exam-Related Topics > SoA/CAS Preliminary Exams > Exam PA: Predictive Analytics
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions


Reply
 
Thread Tools Search this Thread Display Modes
  #311  
Old Today, 10:45 AM
Whoaminoneofyourbusiness's Avatar
Whoaminoneofyourbusiness Whoaminoneofyourbusiness is offline
Member
SOA
 
Join Date: Jan 2017
Location: The Grand Tournament
Studying for PA
Posts: 219
Default

Going to be hard to figure out what to actually do on the exam since the sample project acknowledges that their sample didn't consider a 5 hour timeframe. I guess to start I'll just stick with one method per question and not calibrate parameters I don't need to (complexity parameter, mtry)?

Looks like the sample project must have taken 10-15 hours to do... dunno why they didn't provide a sample solution with exam conditions in mind alongside.
__________________
Spoiler:
Reply With Quote
  #312  
Old Today, 10:55 AM
kasem kasem is offline
SOA
 
Join Date: Apr 2018
Posts: 6
Default

Anyone is going to apply regularization technique in the 5-hour exam? Is it worth to go through material? It seems not practically use for short-term project.
Reply With Quote
  #313  
Old Today, 12:08 PM
jts75 jts75 is offline
Member
SOA
 
Join Date: May 2018
College: Oklahoma State
Posts: 31
Default

Quote:
Originally Posted by Whoaminoneofyourbusiness View Post
Going to be hard to figure out what to actually do on the exam since the sample project acknowledges that their sample didn't consider a 5 hour timeframe. I guess to start I'll just stick with one method per question and not calibrate parameters I don't need to (complexity parameter, mtry)?

Looks like the sample project must have taken 10-15 hours to do... dunno why they didn't provide a sample solution with exam conditions in mind alongside.
I am even more hopeful after seeing the sample solution. They hardly provided details about why those chose methods and didn't provide much detail about their process. Basically it seems like you need to pick a few models that make sense and choose the one with the best statistic you used to compare the models.
Reply With Quote
  #314  
Old Today, 06:29 PM
Demo99 Demo99 is offline
Member
SOA
 
Join Date: Jan 2013
Posts: 133
Default

At this point, how much time does everyone think they'll spend on coding vs writing? Probably 2 hours coding, 3 hours writing for me, I suck at memo's.
Reply With Quote
  #315  
Old Today, 10:07 PM
Squeenasaurus Squeenasaurus is online now
Member
SOA
 
Join Date: Jul 2016
College: Illinois State University
Favorite beer: Lagunitas
Posts: 63
Default

Is there an error on slide 91 of module 6? Should the alpha terms be switched around on the elastic net regression equation? In the RMD file, it says an alpha of 0 implies ridge regression and an alpha of 1 implies lasso. But if you plug in 0 for alpha in the equation on slide 91, you clearly get lasso regression and if you plug in 1 you get ridge regression. Am I missing something?
Reply With Quote
  #316  
Old Today, 10:24 PM
Squeenasaurus Squeenasaurus is online now
Member
SOA
 
Join Date: Jul 2016
College: Illinois State University
Favorite beer: Lagunitas
Posts: 63
Default

Quote:
Originally Posted by Squeenasaurus View Post
Is there an error on slide 91 of module 6? Should the alpha terms be switched around on the elastic net regression equation? In the RMD file, it says an alpha of 0 implies ridge regression and an alpha of 1 implies lasso. But if you plug in 0 for alpha in the equation on slide 91, you clearly get lasso regression and if you plug in 1 you get ridge regression. Am I missing something?
It is an error. It's corrected on slide 100.
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 10:25 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 0.18922 seconds with 9 queries