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
  #241  
Old 09-25-2018, 12:07 PM
Whoaminoneofyourbusiness's Avatar
Whoaminoneofyourbusiness Whoaminoneofyourbusiness is offline
Member
SOA
 
Join Date: Jan 2017
Location: The Grand Tournament
Studying for FAP
Posts: 334
Default

Pretty sure there's a tab on mod9 that will eventually link to a sample solution but it hasn't had a link added yet
__________________
Spoiler:
Reply With Quote
  #242  
Old 09-25-2018, 12:40 PM
ARodOmaha ARodOmaha is offline
Member
SOA
 
Join Date: May 2016
Location: Omaha, NE
Studying for MFE
College: University of Nebraska (alma mater)
Favorite beer: Captain Morgan
Posts: 183
Default

The sample solution is now available on Slide 21 of Module 9.

Quote: "For this sample project, the creators did not consider the fact that only five hours would be available."

Do we need to include analysis for GLM, classification trees, and random forests in our project like they did? I'm assuming they want us to experiment with different models, but not sure how in-depth we need to get.
__________________
P FM MLC MFE C PA FAP APC
Reply With Quote
  #243  
Old 09-25-2018, 01:31 PM
Adapt and Chill's Avatar
Adapt and Chill Adapt and Chill is online now
Member
SOA AAA
 
Join Date: Sep 2017
College: Davidson College
Posts: 131
Default

Quote:
Originally Posted by Whoaminoneofyourbusiness View Post
Anyone know if there's a package we can use to create confusion matrices? Going through the sample code for mod7 and it's using one of the packages not on the list :/
caret has a confusionMatrix function. you just need to pass the predicted results and actual results as arguments. both need to be factors.

confusionMatrix(prediction, actualresults)
Reply With Quote
  #244  
Old 09-25-2018, 01:51 PM
Whoaminoneofyourbusiness's Avatar
Whoaminoneofyourbusiness Whoaminoneofyourbusiness is offline
Member
SOA
 
Join Date: Jan 2017
Location: The Grand Tournament
Studying for FAP
Posts: 334
Default

Got it, thanks. Looks like I needed to load caret manually for that one
__________________
Spoiler:
Reply With Quote
  #245  
Old 09-25-2018, 03:55 PM
Abelian Grape's Avatar
Abelian Grape Abelian Grape is offline
Meme-ber                         Meme-ber
CAS
 
Join Date: Jul 2014
Favorite beer: Allagash Curieux
Posts: 42,069
Default

np
__________________
Quote:
Originally Posted by Pension.Mathematics View Post
what's your problem man?
Reply With Quote
  #246  
Old 09-26-2018, 10:36 AM
Whoaminoneofyourbusiness's Avatar
Whoaminoneofyourbusiness Whoaminoneofyourbusiness is offline
Member
SOA
 
Join Date: Jan 2017
Location: The Grand Tournament
Studying for FAP
Posts: 334
Default

Quote:
Originally Posted by Pension.Mathematics View Post
what's your problem man?
a
__________________
Spoiler:
Reply With Quote
  #247  
Old 09-28-2018, 01:12 PM
cucujay cucujay is offline
 
Join Date: Jan 2011
Posts: 19
Default

On the syllabus, it says "There are four texts required for the course.", anybody using these texts to study or just solely depending on the Modules?
__________________
P FM MLC MFE STAM
VEE FAP PA
Reply With Quote
  #248  
Old 09-28-2018, 01:16 PM
DjPim's Avatar
DjPim DjPim is offline
Member
SOA
 
Join Date: Nov 2015
Location: SoCal
Posts: 332
Default

Quote:
Originally Posted by ARodOmaha View Post
The sample solution is now available on Slide 21 of Module 9.

Quote: "For this sample project, the creators did not consider the fact that only five hours would be available."

Do we need to include analysis for GLM, classification trees, and random forests in our project like they did? I'm assuming they want us to experiment with different models, but not sure how in-depth we need to get.
There's a part in mod 6 where it mentions that 5 hours won't be enough time to check and optimize much, so they expect us to choose one method a priori and defend our choice. I suppose if you're quick enough and do two at the same time, you could just submit whichever one you find to be better, but I'll probably just stick to one.

Not sure if that's exactly your question, but that's how I understood it.
__________________
Quote:
Originally Posted by Dr T Non-Fan View Post
"Cali" SMH.
Reply With Quote
  #249  
Old 09-30-2018, 10:33 PM
DrWillKirby's Avatar
DrWillKirby DrWillKirby is offline
Member
SOA
 
Join Date: Oct 2011
Posts: 943
Default

Quote:
Originally Posted by psynergist View Post
Thanks. Only twice a year
Reply With Quote
  #250  
Old 10-03-2018, 12:19 AM
Squeenasaurus Squeenasaurus is offline
Member
SOA
 
Join Date: Jul 2016
College: Illinois State University
Favorite beer: Lagunitas
Posts: 134
Default

I have a question on the granularity section in Module 4. Maybe I'm just not used to the terminology yet but they say "As long as you have the lowest level of granularity in your data to begin with, you can always create the higher granularity levels later down the track." Don't they mean if you have high granularity you can move to low granularity? If I have data with a bunch of people's address (highly granular), I can group you by which state you live in (low granularity).
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 02:39 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 0.34060 seconds with 10 queries