Calc'ing an EBAR

[BG5150]

Years ago, I created a spreadsheet that calculated an EBAR for a participant given the several items in the equation. 

I cannot find that spreadsheet ANYWHERE (grrrr). It worked perfectly with the UP-84 table.

The element I'm finding tricky is the APR to use.  I know I outputted a table from Relius.  But I forget which one it was.  Whether it was the APR table (maybe it gave like 3 interest rates: 7.5, 8. 8.5) of just the UP-84 table itself and I then calc'd the APR.

Trouble is, I'm not using Relius anymore.  I'm using FT William and that doesn't give access to the underlying tables.

Can someone point me to EITHER:

The APR Table for Up-84?

The equation to calculate the APR itself?

[Bri]

you mean, like, here are the q's and you can take it from there?  i found this on the SOA site....

https://mort.soa.org/ViewTable.aspx?&TableIdentity=831

[austin3515]

I recently did a major upgrade to my "Calculator" and downloaded all of the tables from Relius.  7.5, 8 and 8.5 post retirement, and every "approved" table (so its like 21 columns or something.  I share it here for the good of the community.  Nothing proprietary about mortality tables! 

I can tell you I spent a LOT of time verifying accuracy, etc. before I added it to my calculator. And I make it a habit of using my tool as a starting point, and then verifying it matches the software as a finalizing step (I now use FT William).

So I can tell you their tight but you need to prove on your own before you start using them because I accept zero responsibility for their accuracy.  i.e., don't just take my word for it. 

All APRs (7.5, 8 and 8.5).csv

[austin3515]

1 hour ago, Bri said:

you mean, like, here are the q's and you can take it from there?  i found this on the SOA site....

https://mort.soa.org/ViewTable.aspx?&TableIdentity=831

This is not the APR's though, which is what is needed.  Mine includes the APR's.

[Bri]

oh absolutely, that'll be easier than setting up the commutation functions from scratch (which is why I pointed to a table of q values, depending on how much "first principles" were going to be utilized.)

[austin3515]

Oh yes indeed.  Because if the median of the averages exceeds the regression analysis of q with x being equal to z. that's the other way to do it 🤣🤣🤣

[BG5150]

57 minutes ago, austin3515 said:

I recently did a major upgrade to my "Calculator" and downloaded all of the tables from Relius.  7.5, 8 and 8.5 post retirement, and every "approved" table (so its like 21 columns or something.  I share it here for the good of the community.  Nothing proprietary about mortality tables! 

I can tell you I spent a LOT of time verifying accuracy, etc. before I added it to my calculator. And I make it a habit of using my tool as a starting point, and then verifying it matches the software as a finalizing step (I now use FT William).

So I can tell you their tight but you need to prove on your own before you start using them because I accept zero responsibility for their accuracy.  i.e., don't just take my word for it. 

All APRs (7.5, 8 and 8.5).csv 10.27 kB · 0 downloads

Thanks so much!  This is perfect.

But I have another question.  NRA in this plan is 59.5.  Do I use the APR for 59 or 60?

(I'm gonna try to back into it to see which one matches my software, but I figured I'd ask.

[mming]

From what I've seen, most TPAs use the UP-84 table at 8.5%, which would make one think that combo generally gives the best new comparability results when the owners are among the oldest participants and the NRA is 65.  Has anyone found any patterns where other combos work better for certain ages, demographics, etc. that the IRS probably wouldn't object to?  Also curious as to whether anyone uses more recent mortality tables than UP-84 for this purpose.

[austin3515]

An super good actuary told me that the best approach is individual annuity female 83, 8.5 pre, and 7.5 post.

the 8.5 pre is to get the most bang for the buck on the age disparity between the young and the old.

iaf 83 and 7.5 post was to get the lowest possible EBARS because imputing disparity is more effective at the lower rates (just like it is in a regular integration calc— I.e. you can double anything below 5.7%).