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

DW Simpson
Actuarial Jobs

Visit our site for the most up to date jobs for actuaries.

Actuarial Salary Surveys
Property & Casualty, Health, Life, Pension and Non-Tradtional Jobs.

Actuarial Meeting Schedule
Browse this year's meetings and which recruiters will attend.

Contact DW Simpson
Have a question?
Let's talk.
You'll be glad you did.


Reply
 
Thread Tools Search this Thread Display Modes
  #151  
Old 04-30-2019, 06:18 PM
Squeenasaurus Squeenasaurus is offline
Member
SOA
 
Join Date: Jul 2016
College: Illinois State University
Favorite beer: Lagunitas
Posts: 185
Default

Quote:
Originally Posted by avocado View Post
it's 0.9888, not 0.988
Where'd 0.9888 come from? The exponentiated coefficient of AGE is 0.98764
Reply With Quote
  #152  
Old 05-01-2019, 02:58 PM
LyActuary LyActuary is offline
Member
SOA
 
Join Date: Sep 2017
Location: Rochester, NY
College: University of Rochester
Posts: 102
Default

Gonna start studying for this one after Tuesday's LTAM. I've read through the modules once last year. What is more important, basic data exploration or applying models? Any advice?
Reply With Quote
  #153  
Old 05-01-2019, 04:26 PM
DjPim's Avatar
DjPim DjPim is offline
Member
SOA
 
Join Date: Nov 2015
Location: SoCal
Posts: 713
Default

Quote:
Originally Posted by LyActuary View Post
Gonna start studying for this one after Tuesday's LTAM. I've read through the modules once last year. What is more important, basic data exploration or applying models? Any advice?
What do you mean by applying models?

Data exploration is important, it is one of the 4 main questions / points that is graded. You should be able to produce some graphs, check for / deal with outliers or missing data, and create any new features you think may be useful.

When creating / modifying the code for making the models asked for, you should be able to achieve or interpret the variable selection (depending on method), tune hyper-parameters, assess goodness of fit, interpret high-level results.

Does that help at all?
__________________
Quote:
Originally Posted by Dr T Non-Fan View Post
"Cali" SMH.
Reply With Quote
  #154  
Old 05-02-2019, 11:06 AM
LyActuary LyActuary is offline
Member
SOA
 
Join Date: Sep 2017
Location: Rochester, NY
College: University of Rochester
Posts: 102
Default

Quote:
Originally Posted by DjPim View Post
What do you mean by applying models?

Data exploration is important, it is one of the 4 main questions / points that is graded. You should be able to produce some graphs, check for / deal with outliers or missing data, and create any new features you think may be useful.

When creating / modifying the code for making the models asked for, you should be able to achieve or interpret the variable selection (depending on method), tune hyper-parameters, assess goodness of fit, interpret high-level results.

Does that help at all?
Thank you so much! So to clarify, they gave you all the code for the models and we just need to modify them and do analysis, correct?
Reply With Quote
  #155  
Old 05-02-2019, 11:14 AM
DjPim's Avatar
DjPim DjPim is offline
Member
SOA
 
Join Date: Nov 2015
Location: SoCal
Posts: 713
Default

Quote:
Originally Posted by LyActuary View Post
Thank you so much! So to clarify, they gave you all the code for the models and we just need to modify them and do analysis, correct?
To some extent, yeah. You can check out December's exam and the latest sample test (I think called Student Success?) They both provide some skeleton code to help get the ball rolling. Some people thought they would get extra points by writing an entirely new model or use a totally different method, but for the most part SOA said that was to their detriment, as the code they provided wasn't intended to be bad or wrong, just helping standardize and direct the workflow to let us spend more time cleaning, tuning, and writing.
__________________
Quote:
Originally Posted by Dr T Non-Fan View Post
"Cali" SMH.
Reply With Quote
  #156  
Old 05-03-2019, 07:31 PM
lindy3730's Avatar
lindy3730 lindy3730 is offline
Member
SOA
 
Join Date: Oct 2016
Location: Los Angeles
Studying for FA
College: UC Berkeley
Posts: 64
Default

Noticed that the module updated some materials. I am studying thru the models the second time and found some of the contents are a little bit different.
__________________
__________________
Prelims: P FM MFE MLC C PA
VEE: Statistics Economics Finance
APC

FAP: 1 2 3 4 IA 6 7 FA
Reply With Quote
  #157  
Old 05-04-2019, 03:19 PM
ActuariallyDecentAtBest ActuariallyDecentAtBest is offline
Member
SOA
 
Join Date: Dec 2016
Posts: 383
Default

Quote:
Originally Posted by lindy3730 View Post
Noticed that the module updated some materials. I am studying thru the models the second time and found some of the contents are a little bit different.
Which module/modules?????
Reply With Quote
  #158  
Old 05-05-2019, 01:45 PM
ActuariallyDecentAtBest ActuariallyDecentAtBest is offline
Member
SOA
 
Join Date: Dec 2016
Posts: 383
Default

Can someone explain the difference between an offset and a weight? The modules are not doing a very good job at explaining either of these concepts.
Reply With Quote
  #159  
Old 05-05-2019, 01:58 PM
TranceBrah's Avatar
TranceBrah TranceBrah is offline
Member
SOA
 
Join Date: Mar 2014
Location: Best Coast
Posts: 238
Default

Quote:
Originally Posted by ActuariallyDecentAtBest View Post
Can someone explain the difference between an offset and a weight? The modules are not doing a very good job at explaining either of these concepts.
Imagine a poisson model where Y is the number of accidents (eg. in december's exam) and you have some exposure variable t (employee hours).

no offset: logY ~ b0 + b1x1
including offset: log(Y/t) ~ b0 + b1x1, rearranging into logY ~ logt + b0 + b1x1

Weights change the estimation algorithm by assigning each predictor an "importance". To know how it works you would need to look into how the code is programmed, but this is probably way out of scope for the exam.
Reply With Quote
  #160  
Old 05-05-2019, 02:22 PM
RiskyBusiness7 RiskyBusiness7 is offline
Member
SOA
 
Join Date: Apr 2018
Posts: 53
Default

Quote:
Originally Posted by ActuariallyDecentAtBest View Post
Which module/modules?????
https://www.soa.org/globalassets/ass...-pa-errata.pdf
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 09:02 AM.


Powered by vBulletin®
Copyright ©2000 - 2020, 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 0.20813 seconds with 10 queries