Actuarial Outpost > SoA Can-Do: Task 5
 Register Blogs Wiki FAQ Calendar Search Today's Posts Mark Forums Read
 FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

#1
08-07-2012, 08:10 AM
 irana Member SOA Join Date: Jan 2010 Location: Canada Studying for Exam C & FAP 7 College: Honour bachelor Posts: 109

For sensitivity testing:
Use the base scenario MTM, Assays and hedging% already chosen from task 3 &4. And changing an assumption to measure the the change in Mean?! right?
#2
08-07-2012, 08:22 AM
 ebeebs Member SOA Join Date: Jul 2008 Location: New England Posts: 2,989

Well, not just the change in mean...

Last edited by ebeebs; 08-07-2012 at 08:27 AM..
#3
10-25-2012, 06:25 PM
 Anitha Desai Member SOA AAA Join Date: Jul 2007 Studying for LRM Posts: 9,846

This one's taking me forever.
I just need to finish this up and work on my summary some more.
Can't wait for tomorrow.
Sigh.
__________________
"YOU HUSSY!!! Do our AO vows mean nothing to you? To post and to quote, to RN and to :ctm: for better or worse? Till logout do us part so help you Tom? " - Inconceivable
#4
10-26-2012, 10:33 AM
 mathmajor Member SOA AAA Join Date: Dec 2010 Location: Nowhere in particular Studying for Japanese College: B.S. Applied Math Favorite beer: La Croix Grapefruit Posts: 9,369

Show the effect on everything you think that matters. In particular, I would include your scenario ranking criterion at the very least. You may want to include other measures also, at your discretion. Explain the effect of those parameters on the measures. Why is that happening?
__________________
FSA
#5
10-29-2012, 09:45 AM
 Gedankenexperiment Member Join Date: Dec 2010 Posts: 443

Quote:
 Originally Posted by ebeebs Well, not just the change in mean...
Yeah, positive and negative effects on the mean basically cancel each other out in most cases. I think you want to focus on other things being affected.
__________________
Death is the only decrement.
#6
10-29-2012, 11:32 AM
 mathmajor Member SOA AAA Join Date: Dec 2010 Location: Nowhere in particular Studying for Japanese College: B.S. Applied Math Favorite beer: La Croix Grapefruit Posts: 9,369

I provided a table of outcomes including the base outcome and about five adverse outcomes. For each, I provided every stat modeled, and gave an overview of "what's going on."

Important to remember: suppose in your scenario the mine does not open 35% of the time. There is only one expected profit number for a third of your outcomes.

>effect on mean
>effect on variance and why (usually because the mean drops toward the bulk of outcomes)
>effect on VaR (do mine-open scenarios get worse? mine-closed? both?)
>effect on CTE (usually we have tail events getting more severe)
>when might you re-think opening the mine? how bad does it have to get? is that plausible?
__________________
FSA
#7
11-18-2012, 02:36 PM
 ashley_n_h Member Join Date: Apr 2007 Posts: 125

Did you do this for every assumption? I'm worried about putting in too many tables/graphs..
#8
11-18-2012, 05:54 PM
 waterborne Member SOA Join Date: Nov 2010 Posts: 1,515

Quote:
 Originally Posted by ashley_n_h Did you do this for every assumption? I'm worried about putting in too many tables/graphs..
I did. It's fairly easy to summarize in one table.
#9
11-20-2012, 12:29 PM
 mathmajor Member SOA AAA Join Date: Dec 2010 Location: Nowhere in particular Studying for Japanese College: B.S. Applied Math Favorite beer: La Croix Grapefruit Posts: 9,369

Quote:
 Originally Posted by ashley_n_h Did you do this for every assumption? I'm worried about putting in too many tables/graphs..
Yes. I had one table for each assumption. The table doesn't have to be enormous. This was my longest answer.
__________________
FSA
#10
12-17-2012, 12:15 AM
 its_me Member Join Date: Jan 2003 Location: AO Posts: 2,155

Quote:
 Originally Posted by mathmajor I provided a table of outcomes including the base outcome and about five adverse outcomes. For each, I provided every stat modeled, and gave an overview of "what's going on." Important to remember: suppose in your scenario the mine does not open 35% of the time. There is only one expected profit number for a third of your outcomes. >effect on mean >effect on variance and why (usually because the mean drops toward the bulk of outcomes) >effect on VaR (do mine-open scenarios get worse? mine-closed? both?) >effect on CTE (usually we have tail events getting more severe) >effect on your ranking criteria >when might you re-think opening the mine? how bad does it have to get? is that plausible?
Effect on all of these??

Is it not enough to test for all assumptions and see the effect on Mean & the Metric from Task 2??

Do I need low/moderate/severe fluctuations per assumption - so 6 sensitivities per assumption ? or Is it enough with one positive & one negative movement per assumption?
__________________
Everyone dreams. Some people are just more active participants.

 Tags can do, can-do