Actuarial Outpost > CAS Exam 8 Fall 2018 Progress Thread
 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

#851
Today, 01:00 PM
 SkolChicago Member CAS Join Date: Apr 2015 Posts: 425

Is anyone bothering to spend time studying the Alaska-specific stuff of the NCCI retro manual?
__________________
ACAS 7 8 9
#852
Today, 02:33 PM
 hjacjswo Member Join Date: Dec 2009 College: About to be a super senior Posts: 1,244

Quote:
 Originally Posted by SkolChicago Is anyone bothering to spend time studying the Alaska-specific stuff of the NCCI retro manual?
Whats there to study other than familiarizing yourself with it?
#853
Today, 02:46 PM
 hjacjswo Member Join Date: Dec 2009 College: About to be a super senior Posts: 1,244

Quote:
 Originally Posted by sticks1839 The bolded is incorrect. You use the ICRLL procedure with non-limited table M to approximate the smaller charges for the distribution limited by k. After using this procedure, you still have to add k to the charge. Table L calculates the full charge including the LER for the per-ocurrence limit directly. For example the charge as r -> infinity for table L is k=LER. The charge for table M as r -> infinity is 0.
Okay, thanks. So,

- Table M: The aggregate excess charge/saving based on unlimited claims (No per-occurrence limit / unlimited occurrences)
- Limited Table M: The aggregate excess charge/saving based on limited claims (per-occurrence limit / limited occurrences)
- Table L: The aggregate excess charge/saving + the per-occurrence excess charge/saving based on limited claims (per-occurrence limit / limited occurrences)

To calculate the total loss amounts for the insurer (the one that provides the per-occurrence and the aggregate coverage for the insured), you have to calculate:
expected per-occurrence excess losses + expected excess limited aggregagte losses

Using Limited Table M: k*E(A) + (Limited Table M Charge)*E(limited A)
Using Table L: (Table L Charge)*E(A)

So, you can only apply Limited Table M charge to expected LIMITED losses and then calculate expected per-occurrence excess losses separately, while you can just apply Table L charge to expected UNLIMITED losses to calculate all.

#854
Today, 04:31 PM
 0001 Member CAS Join Date: Sep 2016 Posts: 71

Anyone bothering to learn panjers?
#855
Today, 05:01 PM

Quote:
 Originally Posted by 0001 Anyone bothering to learn panjers?
I'm able to solve these problems from first principles, therefore I won't bother to memorize a formula that makes no sense to me intuitively
__________________
ACAS 7 8 9
#856
Today, 05:20 PM
 Golfer32 Member Non-Actuary Join Date: Apr 2015 Posts: 136

Quote:
 Originally Posted by 0001 Anyone bothering to learn panjers?
Tried and failed a few times. I can't logically piece together the recursive steps. If anyone has a way they think about it that helps them, I'd love to hear it.
#857
Today, 07:13 PM
 SkolChicago Member CAS Join Date: Apr 2015 Posts: 425

Quote:
 Originally Posted by hjacjswo Whats there to study other than familiarizing yourself with it?
Good point.
__________________
ACAS 7 8 9
#858
Today, 07:14 PM
 SkolChicago Member CAS Join Date: Apr 2015 Posts: 425

Was doing a FC and offset came up, so I did some googling to refresh and found this paper on using offset terms in property catastrophe models, in case anyone is interested. Not sure how useful it is for the exam, but the source is very brief on offset terms in my opinion. I plan to read this in more detail after the exam is over.

https://pdfs.semanticscholar.org/c93...5363497370.pdf
__________________
ACAS 7 8 9

 Tags hawaii