Mortality Table - IAM 83 Proj to 2000 with scale G

[flosfur]

Is there an official publication of the IAM 83 Proj to 2000 with scale G mortality table which is universally adopted?

I ask because I am looking at the projected tables from 2 nationally marketed valuation softwares and they are different. What's more, in one vendor's version of the male table, the q at age 41 is almost twice the q at age 40 and q41 & q42 are higher than the qs in the base table IAM83M - which is clearly wrong!

I also compared these tables to the similarly projected tables titled 1996 US Annuity 2000 ...., both the Basic and the 10% adjusted, from the SOA's website and neither vendor's tables match the SOA's version.

[Gary]

I would have thought the SOA would have the table you are searching for.

[david rigby]

SOA table Manager. http://library.soa.org:8080/xtbml/jsp/index.jsp

Click on "Catalog of all available titles"

Scale G is tables 908 and 909.

83 IAM Basic, Female = table 823 83 IAM Basic, Male = table 824

83 IAM, Female = table 829 83 IAM, Male = table 830

If you need to know how to proceed, or which table is most suitable to your situation, or how to apply the projection scale, I am available for hire.

[Guest Doug Goelz]

SOA table Manager. http://library.soa.org:8080/xtbml/jsp/index.jsp

Click on "Catalog of all available titles"

Scale G is tables 908 and 909.

83 IAM Basic, Female = table 823 83 IAM Basic, Male = table 824

83 IAM, Female = table 829 83 IAM, Male = table 830

If you need to know how to proceed, or which table is most suitable to your situation, or how to apply the projection scale, I am available for hire.

I don't believe there is any official version of the projected 1983 IAM table that you mentioned. However, taking the regular 1983 IAM table and projecting the qx's 17 years with Scale G would be the approach I would take. I attached an Excel file of the projected rates. How do these compare with your national software tables?

1983_IAM_Table_Projected_17_Years_with_Scale_G.xls

[AndyH]

And I thought you (pax) were too busy attending all those BOSTON RED SOX minor league affiliate baseball games during the summer.

[david rigby]

And I thought you (pax) were too busy attending all those BOSTON RED SOX minor league affiliate baseball games during the summer.

Ouch! Go Braves!

[flosfur]

I don't believe there is any official version of the projected 1983 IAM table that you mentioned. However, taking the regular 1983 IAM table and projecting the qx's 17 years with Scale G would be the approach I would take. I attached an Excel file of the projected rates. How do these compare with your national software tables?

Thanks.

The software vendors' projected qs are different than yours. Some are exactly the same, some differ on the 6th decimal and some on 5th & 6th decimal. For example, for females

1000 times your q15=167 vs 159 and 172; 1000 times your q16=181 vs 170 and 184

1000 times your q23=283 vs 286 and 285; 1000 times your q26=327 vs 338 and 340

1000 times your q30=369 vs 388 and 356; 1000 times your q40=504 vs 508 and 503

So they are all over the place.

[flosfur]

SOA table Manager. http://library.soa.org:8080/xtbml/jsp/index.jsp

Click on "Catalog of all available titles"

Scale G is tables 908 and 909.

83 IAM Basic, Female = table 823 83 IAM Basic, Male = table 824

83 IAM, Female = table 829 83 IAM, Male = table 830

If you need to know how to proceed, or which table is most suitable to your situation, or how to apply the projection scale, I am available for hire.

I know how to do it. But I am just trying to get a universally agreeable table. Now I have tables from 3 sources and they are all different.

Some clown put "IAM 83 Female Proj to 2000 with scale G" as the mortality table for Actuarial Equivalent in the plan document I am looking at, so I want to make sure I am using the right table. But it looks like there is no such thing in this case.

[david rigby]

Some clown put "IAM 83 Female Proj to 2000 with scale G" as the mortality table for Actuarial Equivalent in the plan document I am looking at, so I want to make sure I am using the right table.

Have you considered contacting the clown, or perhaps the prior actuary?

But it looks like there is no such thing in this case.

Sure there is, exactly what the defintion states. Don't know if it was done correctly, but we should not assume a table is invalid just because it is not "standard". The purpose of (any) scale is to permit the modification of any table, thus creating a new table.

[flosfur]

Have you considered contacting the clown, or perhaps the prior actuary?

The clown & the actuary are one and the same. But what would contacting him will achieve?

The problem is getting the tables projected with the same scale (G) by various sources to be the same.

But it looks like there is no such thing in this case.

Sure there is, exactly what the defintion states. Don't know if it was done correctly, but we should not assume a table is invalid just because it is not "standard". The purpose of (any) scale is to permit the modification of any table, thus creating a new table.

"Sure there is"?

Shouldn't 2+2 equal 4, no matter who computes it or what instrument is used to compute it. But here it seems not to be the case. As mentioned above, IAM 83 table projected to 2000 with Scale G from 3 sources don't have the same qs and hence 2+2 has 3 different values!

If the projected qs had been put in the plan document then there will be no problem - right or wrong, they are in the plan document and they are what they are and the annuity factors based on them would be the same no matter who computed them, i.e. definitely determinable!

[Mike Preston]

Nobody puts q's into documents. Not of fairly well established tables, anyway. I think it is commonly known that I was the source of the original calculation "way back when." Not that it was particularly rocket science. I just used the article published in the Transactions of the Society (green books, back then) which detailed both the projection factors and the methodology for using them. I had my calculations peer reviewed by at least 3 other actuaries who all agreed with my projections. Alas, I have lost the spreadsheet I used to do the projections. However, the APR's for common ages (55,62,65) are well known to those of us who have used the table for many, many years. You can probably find confirmation of those APR's on BenefitsLink if you look hard enough.

Or, you could just ask.

[flosfur]

Nobody puts q's into documents. Not of fairly well established tables, anyway. I think it is commonly known that I was the source of the original calculation "way back when." Not that it was particularly rocket science. ....

This doesn't explain why two national software vendors have different qs and why neither of them agree with Doug Goelz's projected qs attached to his message above. And one vendor's q41 & q42 are higher than the base qs.

.....

Or, you could just ask.

I thought that's exactly what I did!

[Mike Preston]

q's != APR's

[david rigby]

... IAM 83 table projected to 2000 with Scale G from 3 sources don't have the same qs and hence 2+2 has 3 different values!

Incorrect. It means somebody did it wrong, not that the table is invalid.

[Mike Preston]

Does anybody have a copy of the Transactions where the methodology was described? I seem to recall it wasn't straightforward. That is, without the writeup, if all you have to look at is the projection factors, there is a high likelihood that you would not duplicate the intended use of Projection Scale G.

[Guest Doug Goelz]

Does anybody have a copy of the Transactions where the methodology was described? I seem to recall it wasn't straightforward. That is, without the writeup, if all you have to look at is the projection factors, there is a high likelihood that you would not duplicate the intended use of Projection Scale G.

Is there any chance your projected 1983 IAM table is really the Annuity 2000 table? See attachment from SOA that talks about how the Annuity 2000 table was developed from projecting the 1983 table with Scale G.

I didn't read the whole note, but perhaps there are other clues in it that you will find useful.

Annuity_2000_Mortality_Table_Notes.pdf

[Mike Preston]

Thanks, Doug. I'll try to review that when an opportunity presents itself. At first glance, though, it appears that the projection is based on 100% of Scale G for males and 50% of Scale G for females. I think that is inconsistent with the original notes as published in the Transactions. FWIW.