Actuarial Outpost
 
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Life
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions


Upload your resume securely at https://www.dwsimpson.com
to be contacted when our jobs meet your skills and objectives.


Reply
 
Thread Tools Display Modes
  #1  
Old 10-23-2012, 08:42 PM
epshiple epshiple is offline
Member
SOA
 
Join Date: Mar 2012
Location: Phoenix
Studying for MLC
College: Arizona State University
Posts: 33
Default Joint Mortality Equivalency

I am working on a project right now and have hit a wall on how to proceed. Product I am working on is a second to die product.

I have the qx's (probability of death in year x) for two single policies. I am wanting to show the equivalent qx of a joint policy comprised of the two individuals.

Note - I am not looking for the probability of the second death happening by year x, but rather on a year by year basis. So I want the probability that in year x the policy ends.

My best guess at the moment is to multiply the conditional probabilites of death for each person given that the other is dead.

Example - Person A and Person B

(ADead*BDead) / [(ADead*BDead)+(ALives*BDead)]

(BDead*ADead) / [(BDead*ADead) + (BLives*ADead)]

I take these two products and multiply them together and get what looks like a logical curve for the joint policy. It is always outside of the single policy curves, and it also meets up with the second curve once the first curve hits ultimate probability of death. Is there something I am missing in this computation?

Last edited by epshiple; 10-23-2012 at 09:03 PM..
Reply With Quote
  #2  
Old 10-24-2012, 05:29 AM
george24 george24 is offline
Member
SOA
 
Join Date: Aug 2011
Studying for nothing
Posts: 264
Default

why not just the sum of 3 pieces:

both are alive, and then both die that year
A is alive then dies, B is already dead
B is alive then dies, A is already dead

that's assuming independence, but you'd probably want to factor in Lonely Heart Syndrome, etc.
Reply With Quote
  #3  
Old 10-24-2012, 08:30 AM
JMO's Avatar
JMO JMO is offline
Carol Marler
 
Join Date: Sep 2001
Location: Back home again in Indiana
Studying for Nothing actuarial.
Posts: 30,855
Default

See article on page 4:
http://www.soa.org/library/newslette...h/act7803.aspx

And letters to the editor in this subsequent issue.
http://www.soa.org/library/newslette...e/act7806.aspx
__________________
Carol Marler, FSA, MAAA, A Former Actuary
Dedicated to Retirement
Just My Opinion (Although this statement is my opinion, and I am still listed as an actuary, it's still not a statement of actuarial opinion, and you really shouldn't rely on it.)

Last edited by JMO; 10-24-2012 at 08:36 AM..
Reply With Quote
  #4  
Old 10-24-2012, 09:24 AM
MathGeek92 MathGeek92 is offline
Member
 
Join Date: Jul 2004
Posts: 460
Default

or.. just find out the Pxy and Pxy+1 and subtract the two... where Pxy is the probability that at least one is alive
Reply With Quote
  #5  
Old 10-24-2012, 04:33 PM
epshiple epshiple is offline
Member
SOA
 
Join Date: Mar 2012
Location: Phoenix
Studying for MLC
College: Arizona State University
Posts: 33
Default

I have taken a look at Frasier's method. Does this method work only with a joint policy comprised of identical age single policies?


If it does work among combinations of ages...

I have been transferring my results onto a graph to show the relationships of the two single policy mortality curves vs. the joint curve for the two and it just does not make sense to me that my joint line surpasses my single line at certain periods of time.

In other words, if I have person X, how is it that his probability of death is lower than person X joined with person Y for a joint policy. In the way I am looking at it, adding person Y should do nothing but decrease the likelihood of the policy ending, because now we need two people to die rather than just one. I have attached an excel sheet summarizing as I am not the best at describing the situation.

I understand the outside forces like broken heart syndrome, but my calculation should be treating it as if the two are independent. I want to understand that part before I move to bringing in dependencies between the two.
Attached Files
File Type: xlsx Outpost Mortality.xlsx (47.5 KB, 79 views)
Reply With Quote
  #6  
Old 10-24-2012, 07:10 PM
Double High C's Avatar
Double High C Double High C is offline
Site Supporter
Site Supporter
SOA AAA
 
Join Date: Sep 2001
Location: atop the 5th line above the staff
Posts: 11,465
Default

Quote:
Originally Posted by epshiple View Post
I have taken a look at Frasier's method. Does this method work only with a joint policy comprised of identical age single policies?
No. It works in general, for 2 independent lives.

I didn't look at your file, but it was stated succintly in the quote below, but I added one clarification that may be your missing link:


Quote:
Originally Posted by MathGeek92 View Post
or.. just find out the Pxy and Pxy+1 and subtract the two... where Pxy is the conditional probability that at least one is alive given that AT LEAST ONE - but maybe not two - was alive at the beginning of the prior year
(And of course, they were both alive at the issyue date.)

ICHP

All Fraser's method does is take a conditional probability, i.e. a weighted average of the probabilities of the three statuses.

Last edited by Double High C; 10-24-2012 at 07:16 PM..
Reply With Quote
  #7  
Old 10-24-2012, 09:25 PM
JoJo JoJo is offline
Member
 
Join Date: Jan 2002
Posts: 1,014
Default

An interesting read is the study on simultaneous deaths.

http://www.soa.org/search.aspx?searchterm="simultaneous%20death"

This study showed a whopping annual rate of .04 per 1000. I've seen lots of early duration COIs on survivorship policies that are way underpriced if the rate is really .04 per 1000.
Reply With Quote
Reply

Thread Tools
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 08:07 PM.


Powered by vBulletin®
Copyright ©2000 - 2014, 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 0.35759 seconds with 8 queries