Actuarial Outpost
 
Go Back   Actuarial Outpost > Exams - Please Limit Discussion to Exam-Related Topics > CAS > CAS Exams > Exam 7 - Estimation of Policy Liability & ERM
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

DW Simpson Global Actuarial & Analytics Recruitment
Download our Actuarial Salary Survey
now with state-by-state salary information!


Reply
 
Thread Tools Search this Thread Display Modes
  #651  
Old 03-06-2020, 01:48 PM
AbedNadir's Avatar
AbedNadir AbedNadir is offline
Member
CAS SOA
 
Join Date: Mar 2014
Studying for FCAS
Posts: 2,848
Default

practice question: you are testing a bayesian model that does not include a calendar year trend. The histogram shows your model is biased high. You decide to add a calendar year trend component. Draw what you expect to be the resulting histogram of the posterior distribution for the mean of your new trend parameter.
__________________
Reply With Quote
  #652  
Old 03-06-2020, 02:03 PM
AbedNadir's Avatar
AbedNadir AbedNadir is offline
Member
CAS SOA
 
Join Date: Mar 2014
Studying for FCAS
Posts: 2,848
Default

Quote:
Originally Posted by AbedNadir View Post
Clark shows that the overall variance is lower for CC due to the trade off being better for decreasing parameter variance and increasing process variance. The discussion by Meyer's is about overall variance as you said. So we have Clark advocating for lower overall variance, and Meyer's trying to get more overall variance
it could also be due to the time period difference in which these were written. As Meyer's said, nobody has done much effort to actually validate models against real losses and he wanted to be the first to demonstrate a systematic effort. Clark may have been operating under the assumption that the models are unbiased so tightening up the variability was his goal. Meyer's has a different perspective and with his specific test using histograms and p-p plots, having a small variability estimate turns out to be worse
__________________
Reply With Quote
  #653  
Old 03-06-2020, 03:50 PM
SkolChicago SkolChicago is offline
Member
CAS
 
Join Date: Apr 2015
Posts: 1,245
Default

Well Im back after *checks post history* a month? Jesus. Question on Mack 1994 - more so trying to make sure I have this right:

CF has a section on how individual reserve squared standard errors dont add to the total reserve squared standard error. However, the cookbook has a recipe on using the same lognormal formulas as those for individual accident years (formulas [10]) from the paper). Is it ok to use these formulas on the overall reserve?
__________________
ACAS 7 8 9
Reply With Quote
  #654  
Old 03-06-2020, 04:46 PM
AbedNadir's Avatar
AbedNadir AbedNadir is offline
Member
CAS SOA
 
Join Date: Mar 2014
Studying for FCAS
Posts: 2,848
Default

Quote:
Originally Posted by SkolChicago View Post
Well I’m back after *checks post history* a month? Jesus. Question on Mack 1994 - more so trying to make sure I have this right:

CF has a section on how individual reserve squared standard errors don’t add to the total reserve squared standard error. However, the cookbook has a recipe on using the same lognormal formulas as those for individual accident years (formulas [10]) from the paper). Is it ok to use these formulas on the overall reserve?
yes that's precisely what Meyers does in chapter 3 of his paper
__________________
Reply With Quote
  #655  
Old 03-07-2020, 01:02 AM
Candidate X's Avatar
Candidate X Candidate X is offline
Member
CAS
 
Join Date: Oct 2015
Posts: 310
Default

I believe it's true that the sum of the squares of the individual reserve standard errors don't add to the reserve squared standard error. (See Mack eq 11). However, once you do have the total SE, you should be able to use the same *procedure* (equivalent version of eq. 10) as a step toward calculating confidence intervals for the total reserve estimate. I don't intend to memorize eq. 11, however, since one problem of that type would probably take me more than 4 hours without a spreadsheet to calculate. I'll probably try to remember eq 7, though to apply to an individual accident year
Reply With Quote
  #656  
Old 03-09-2020, 09:27 AM
amp019372's Avatar
amp019372 amp019372 is offline
Member
CAS
 
Join Date: Aug 2017
Location: Columbus, OH
Studying for 7
College: B.S. University of TN, M.S. Ohio State
Favorite beer: Spotted Cow
Posts: 813
Default

I'm now about halfway done with Brehm and going through past problems. TIA seems to be way too detailed for Brehm, but I suppose that's better than not having enough information and skipping things. Brehm definitely feels like my weakest section (probably because a lot of it is qualitative), but overall I'm starting to feel okay about this exam.

Goal is to finish Brehm at 6 weeks out, then start reviewing everything to fill in all the details. Now that I think about it, I finished all the material for exam 6 and started reviewing 5 weeks out, so I guess I'm a little ahead of schedule compared to then. And since this exam does feel a lot like 6, I guess that's a good thing. When I took 6 I definitely was weakest on the SAO section and it feels analogous to how I feel about Brehm at the moment. But I guess that's fine as long as I nail the other stuff down.

50 days left!
__________________
ACAS | 7 | 8 | 9



Reply With Quote
  #657  
Old 03-09-2020, 10:22 AM
SkolChicago SkolChicago is offline
Member
CAS
 
Join Date: Apr 2015
Posts: 1,245
Default

Do y’all think the reserving papers (or any specific ones) are prerequisites for Goldfarb and Brehm? Given the syllabus weight for those I’m thinking about setting some mandatory start date for myself on those papers, like a month out.
__________________
ACAS 7 8 9
Reply With Quote
  #658  
Old 03-09-2020, 10:38 AM
AbedNadir's Avatar
AbedNadir AbedNadir is offline
Member
CAS SOA
 
Join Date: Mar 2014
Studying for FCAS
Posts: 2,848
Default

Quote:
Originally Posted by SkolChicago View Post
Do yall think the reserving papers (or any specific ones) are prerequisites for Goldfarb and Brehm? Given the syllabus weight for those Im thinking about setting some mandatory start date for myself on those papers, like a month out.

An understanding of the valuation process should be beneficial to understand some of the knowledge points in Goldfarb but particular methods don't really matter.

Brehm is related to stochastic methods in the sense that they provide distributions for unpaid losses, and those distributions can be used in an enterprise risk model. Meyer's even starts off his paper by saying that distributions are becoming more important due to an increased application of enterprise risk management techniques to insurance. Finding the connections to reserving methods might help your recall of certain sections of Brehm and understanding what the story is all about but you just need to know the concepts mostly
__________________
Reply With Quote
  #659  
Old 03-09-2020, 11:35 AM
AbedNadir's Avatar
AbedNadir AbedNadir is offline
Member
CAS SOA
 
Join Date: Mar 2014
Studying for FCAS
Posts: 2,848
Default

chapter 6 of Shapland also has numerous clear connections to Brehm if you're looking
__________________
Reply With Quote
  #660  
Old 03-09-2020, 01:41 PM
HuskerCAS HuskerCAS is offline
Member
CAS AAA
 
Join Date: Mar 2011
College: Nebraska
Posts: 508
Default

Quote:
Originally Posted by SkolChicago View Post
Do yall think the reserving papers (or any specific ones) are prerequisites for Goldfarb and Brehm? Given the syllabus weight for those Im thinking about setting some mandatory start date for myself on those papers, like a month out.
One potential "gotcha" is making us calculate a reserve pick for a goldfarb cash flow question.

The hard part would be that it is on a calendar year basis, and the paper that covers reinsurance contracts has examples of converting policy year and accident year. I suppose nothing is stopping them from doing the balance sheet as PY or AY and then giving us everything we need on the premium and discounting side.

I'm not sure if such a calculation is on the syllabus (feels more like an exam 6 or exam 9 topic), but if they wanted to do a synthesis question with Goldfarb that's how they would have to do it.

Relating Brehm and Goldfarb, you cannot simply add lines of business because of the inter-line correlation. That's also an issue for Shapland if you wanted to ODP Bootstrap your way to an Enterprise reserve. Marshall also covers a similar topic with the correlated CoV's.
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 09:10 PM.


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 0.34685 seconds with 10 queries