Actuarial Outpost Generational Mortality Table: 430(h)(3)-1
 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

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

#1
01-15-2008, 04:04 PM
 volva yet Note Contributor Join Date: Feb 2006 Location: Nomadic Studying for GHC/DMAC College: PSU '07 Favorite beer: Oskar Blues Old Chub Scotch Ale Posts: 4,949
Generational Mortality Table: 430(h)(3)-1

This is for anyone who was interested in generating mortality tables based on the Calendar Year 2000 Base Tables by using the AA Scale factors.

The Annuitant table is generated as follows...

Ages < 40: determined using a projection period that runs from the year 2000 until 15 years after the calendar year that contains the valuation date for the plan year.
Ages > 50: determined using a projection period that runs from the year 2000 until 7 years after the calendar year that contains the valuation date for the plan year.
Ages between 40 and 50: curve fit between the two, most likely it is compounded

I found this by trying to reproduce the 2008 Static Annuitant tables. By 430(h)(3)-1(c)(2), one would think that the annuitant table has a projection period +7 for the entire table, but it doesn't. It seems to make sense that this only happens for ages greater than 50 (i.e. where retirement benefits become more likely). Nevertheless, this mixed distribution wasn't explained well in the Proposed Treasury Regulation.
__________________
Exams
How to explain actuarial exams to someone else...

Good Einstein quote - "One had to cram all this stuff into one's mind for the examinations, whether one liked it or not. This coercion had such a deterring effect on me that, after I had passed the final examination, I found the consideration of any scientific problems distasteful to me for an entire year."
#2
01-16-2008, 03:23 PM
 volva yet Note Contributor Join Date: Feb 2006 Location: Nomadic Studying for GHC/DMAC College: PSU '07 Favorite beer: Oskar Blues Old Chub Scotch Ale Posts: 4,949

Male mortality tables are curve fit using a sequence such as: $a_{x+1}=a_x+c_y$, where $y=x-40$ and $x$ is age.

FYI, I have no clue on how to curve fit the data for the female mortality tables.
__________________
Exams
How to explain actuarial exams to someone else...

Good Einstein quote - "One had to cram all this stuff into one's mind for the examinations, whether one liked it or not. This coercion had such a deterring effect on me that, after I had passed the final examination, I found the consideration of any scientific problems distasteful to me for an entire year."
#3
01-22-2008, 02:07 PM
 Rick_G Member Join Date: May 2005 Location: Just south of the Big Chicken Posts: 2,209

Quote:
 Originally Posted by colby2152 Male mortality tables are curve fit using a sequence such as: $a_{x+1}=a_x+c_y$, where $y=x-40$ and $x$ is age. FYI, I have no clue on how to curve fit the data for the female mortality tables.
Someone asked me yesterday about the mortality qx's. I had no idea either, since the IRS did not give all the details in the regulation. After asking around a bit, I got the answer from Alan Turk at AON (he is an actuary I used to work with):

"The basic method is to take the separate RP-2000 annuitant and non-annuitant tables and project them to the valuation date plus 7 years for the annuitant tables, 15 years for non-annuitant tables. These original tables are partials.

The annuitant table starts at age 50 and the non-annuitant table ends at age 70, so you fill in missing ages from the opposing table (i.e. ages 80 to 120 from the annuitant table into the non-annuitant and ages 1 to 40 from the non-annuitant into the annuitant table).

You then apply a second difference smoothing to cross the 10 year gaps. There is an exception for the female annuitant table; non-annuitant rates through age 44 are used and the bridge is between 44 and 50. The combined tables are a weighted average of the annuitant and non-annuitant rates, the weights are published. Then lastly, take 50/50 males and females to get the 417(e) mortality.

The IRS did something else funny. For the base tables to project generationally, the female annuitant table bridge is between 46 and 50, not 44 and 50."
__________________
Rick Groszkiewicz

Now offering online seminars and everything else under the sun for actuarial exams.
EA-2F and EA-2L and EA-1
#4
01-22-2008, 04:36 PM
 volva yet Note Contributor Join Date: Feb 2006 Location: Nomadic Studying for GHC/DMAC College: PSU '07 Favorite beer: Oskar Blues Old Chub Scotch Ale Posts: 4,949

Thanks Rick... I figured this out the hard way! The female tables were very tough, and I gave up on figuring the exact curve fit.
__________________
Exams
How to explain actuarial exams to someone else...

Good Einstein quote - "One had to cram all this stuff into one's mind for the examinations, whether one liked it or not. This coercion had such a deterring effect on me that, after I had passed the final examination, I found the consideration of any scientific problems distasteful to me for an entire year."
#5
02-08-2008, 08:15 PM
 Doug G. Join Date: Sep 2003 Location: Pittsburgh Posts: 29

I just noticed this post. Since somebody asked something similar on BenefitsLink, I decided to upload a couple of files that you may be interested in reviewing. The Excel file is geared toward 2007, but will suffice to show the methodology. It can also be used to project to 2008 or later if desired.
Attached Files
 Smoothing Note on CL Mortality.doc (48.5 KB, 263 views) RP2000 Table Notes - 2007.xls (183.5 KB, 437 views)