Actuarial Outpost
 
Go Back   Actuarial Outpost > Exams - Please Limit Discussion to Exam-Related Topics > SoA > Modules 6-8
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
  #21  
Old 08-04-2012, 11:55 PM
komorgan komorgan is offline
Member
Non-Actuary
 
Join Date: Mar 2012
Posts: 109
Default

Quote:
Originally Posted by ebeebs View Post
What happens if somebody falls and dies? If a worker cuts off his hand? A faulty beam collapses and somebody gets injured? Maybe some environmentalists take them to court over ruining the water supply, etc...

Perhaps there's something you can add to the model to cover legal liabilities...
Interesting. I had always thought they wanted us to just suggest better ways of quantifying the existing assumptions. It never really occurred to me that they expected us to propose new assumptions. Regarding the idea of a legal liabilities assumption, I don't really know of any particular models (either deterministic or stochastic) that estimate legal liabilities. I suppose we could make a "best guess" estimate, based on historical legal liabilities of similar mining companies.

Any thoughts?
Reply With Quote
  #22  
Old 08-05-2012, 12:18 AM
irana irana is offline
Member
SOA
 
Join Date: Jan 2010
Location: Canada
Studying for Exam C & FAP 7
College: Honour bachelor
Posts: 109
Default

Or rather than being so specific you could just mention model documentation, which should be more specific. EOM6 is very helpful.
Reply With Quote
  #23  
Old 08-05-2012, 02:51 PM
ebeebs ebeebs is offline
Member
SOA
 
Join Date: Jul 2008
Location: New England
College: '09 grad
Posts: 2,989
Default

Quote:
Originally Posted by komorgan View Post
Interesting. I had always thought they wanted us to just suggest better ways of quantifying the existing assumptions. It never really occurred to me that they expected us to propose new assumptions. Regarding the idea of a legal liabilities assumption, I don't really know of any particular models (either deterministic or stochastic) that estimate legal liabilities. I suppose we could make a "best guess" estimate, based on historical legal liabilities of similar mining companies.

Any thoughts?
The good thing about this is you don't actually need to build a model or know how to. You just have to make the suggestions. This is what my Task 1 looked like.

Economic
1)Thing to fix
-why
-what would change
-easy or hard to incorporate (ex- interest rates being stochastic would take a lot of work and research)
-how material the change would be (big or small model improvement)
-any further explanation you want
-yes I did this section in bullets because it's so much easier to read

2) same things as above.
3,4, etc...

Non-economic
1,2,3 etc same as above.

They can be both new things to add or fixes on current things in the model.
Reply With Quote
  #24  
Old 08-10-2012, 04:45 PM
IIRC IIRC is offline
Member
 
Join Date: Apr 2009
Posts: 397
Default

Quote:
Originally Posted by ebeebs View Post
The good thing about this is you don't actually need to build a model or know how to. You just have to make the suggestions. This is what my Task 1 looked like.

Economic
1)Thing to fix
-why
-what would change
-easy or hard to incorporate (ex- interest rates being stochastic would take a lot of work and research)
-how material the change would be (big or small model improvement)
-any further explanation you want
-yes I did this section in bullets because it's so much easier to read

2) same things as above.
3,4, etc...

Non-economic
1,2,3 etc same as above.

They can be both new things to add or fixes on current things in the model.
This helps, thanks.

If I suggest something could benefit from being modeled stochastic, how can I (or do I need to) expand on that?

i.e. Exchange. Lots of folks suggest to model it stochastic. Given the model is level for 20 years, it is unrealistic to expect exchange to not change. So it would benefit from some sort of random movement.

Is this enough of an explaination? I don't have a clue about modeling something stochastic. Sorry if I'm vague, kind of scrambling here and have tunnel vision thinking of this FA for too long.

In short, is it enough to say something like, the model might benefit from exchange having a more random element (vs. linear) via an exchange rate model...?

Thanks
Reply With Quote
  #25  
Old 08-10-2012, 04:48 PM
IIRC IIRC is offline
Member
 
Join Date: Apr 2009
Posts: 397
Default

Maybe the model is just too long (20 years). Who can forcast anything that has any degree of volatility for 5, 10, years let alone 20?
Reply With Quote
  #26  
Old 08-10-2012, 05:50 PM
IIRC IIRC is offline
Member
 
Join Date: Apr 2009
Posts: 397
Default

In regards to the exchange rate. Anybody have any ideas how to tie a realistic change into the exchange rate via. the simulated gold prices in light of the negative correlation between USD and Gold?

i.e. when gold is up, USD is stronger, thus the exchange is weaker, right? this might be a plausible change that would have a stochastic quality via. the gold prices...

Thoughts?
Reply With Quote
  #27  
Old 09-28-2012, 08:31 AM
BAIIPRO BAIIPRO is offline
Member
SOA
 
Join Date: Jul 2009
Posts: 260
Default

Did people generally provide a recommendation with their possible improvements?

It says to "offer possible improvements" and the benefits/risks associated with that improvement. But if an improvement is high work with high reward, I can't decide if I'd like to make a recommendation or let that to my imaginary manager...especially given that this is presumably the first task you're doing and you don't have a ton of experience with the model yet.
Reply With Quote
  #28  
Old 09-28-2012, 08:34 AM
BAIIPRO BAIIPRO is offline
Member
SOA
 
Join Date: Jul 2009
Posts: 260
Default

I'm also thinking of listing all of the possible improvements but then summarizing with a list of the improvements, ranked by which I would pursue first, second, third...and so on.
Reply With Quote
  #29  
Old 09-28-2012, 08:41 AM
ebeebs ebeebs is offline
Member
SOA
 
Join Date: Jul 2008
Location: New England
College: '09 grad
Posts: 2,989
Default

Quote:
Originally Posted by BAIIPRO View Post
Did people generally provide a recommendation with their possible improvements?

It says to "offer possible improvements" and the benefits/risks associated with that improvement. But if an improvement is high work with high reward, I can't decide if I'd like to make a recommendation or let that to my imaginary manager...especially given that this is presumably the first task you're doing and you don't have a ton of experience with the model yet.
Offer possible improvements. You don't have to say anything about if you would suggest implementing it.

Quote:
Originally Posted by BAIIPRO View Post
I'm also thinking of listing all of the possible improvements but then summarizing with a list of the improvements, ranked by which I would pursue first, second, third...and so on.
I wouldn't. Just answer exactly what it asks imo.
Reply With Quote
  #30  
Old 09-29-2012, 09:50 AM
BAIIPRO BAIIPRO is offline
Member
SOA
 
Join Date: Jul 2009
Posts: 260
Default

I don't do any modeling in my job, so still kind of figuring out the logistics of it as I go.

Question: If I'm recommending that we stochastically model the investment rate. How would that model incorporate itself into the rest of the Can-Do model? For instance, would I model 100 strings of 80 interest rates (100 scenarios and 80 quarters in each), and then use those as the monthly interest rates for my scenarios?

Or would I model the same number of interest rates, but take some sort of resulting average to use (deterministically) in my Can-Do model?

Obviously the first option would be much more time consuming? Also, how to we answer the question of 'how hard would it be to implement'? I have no idea how much time it would take...only that it would take considerable research and computer resources.
Reply With Quote
Reply

Tags
can do, can-do

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 02:14 AM.


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 0.57654 seconds with 9 queries