

FlashChat  Actuarial Discussion  Preliminary Exams  CAS/SOA Exams  Cyberchat  Around the World  Suggestions 
LongTerm Actuarial Math Old Exam MLC Forum 

Thread Tools  Search this Thread  Display Modes 
#1




Multiple Decrement Probability Question from ASM Manual
This is from ASM MLC Manual, 15th Edition, 4th printing, page 1024, question 50.7.
The question: For a doubledecrement table, you are given: 1.) Decrements from cause 1 are uniformly distributed over each year of age in the associated single decrement table 2.) In the associated single decrement table, 40% of decrements from cause 2 occur at time 0.25 and 60% occur at time 0.6 within each year of age 3.) q'(1) (age 45) = .1 4.) q'(2) (age 45) = .05 Calculate q(2) (age 45) My work assumes starting with one life. Then there should be 1  (.25)(.1) =.975 lives at time 0.25 due to cause 1. Then cause 2 acts on the remaining .975 lives as follows: .975(.4)(.05) = .0195 decrements. So at time 0.25, after both decrements, there are .975  .0195 = .9555 lives remaining. Then at time 0.6, there are .9555(1.35(.1)) = .9220575 lives due to cause 1. Cause 2 acts on those lives at follows: .9220575(0.60)(0.05)=.02766. Therefore q(2) (age 45) = .0195 + .02766 = .04716. The solution does this math: "The exposure after cause 1 is 1.25*(0.1) = .975 at time 0.25 and 1(0.6)(0.1) = .94 at time 0.6. Therefore: q(2)(age 45) = .05*.4*.975 + .05*.6*.94 = .0477. Why is my method flawed? 
#2




Have you considered what happens from time 0.6 to time 1.0?
__________________
"What do you mean I don't have the prerequisites for this class? I've failed it twice before!" "I think that probably clarifies things pretty good by itself." 
#3




Doesn't cause 2 stop impacting the survivors immediately following time 0.6, at which time cause 1 is the only remaining decrement? I must be missing something, but I don't see how time 0.6 to time 1 applies for cause 2.

#4




Apply your approach to 100 lives with the first decrement set to 0. You would get
2 + 98 x .03 decrements. 
#5




The error I see in your approach is in computing the fraction of the 0.9555 alive at time 0.25 that are killed by #1 in the next 0.35. You take that fraction to be 0.35 x 0.1, but that is actually the fraction of a whole person at time 0 that would be killed in that interval. The fraction of a whole person alive at time 0.25 that would be killed in the next 0.35 by #1 would be 0.35 x .1 / [1  0.25 x .1].
Quote:
__________________
Jim Daniel Jim Daniel's Actuarial Seminars www.actuarialseminars.com jimdaniel@actuarialseminars.com 
#6




Thank you Jim. This started to click a few hours after I posted this.

Thread Tools  Search this Thread 
Display Modes  

