Life table for men

[Owner]

Dear sci.stat.consult:
With the help of everyone here steering me toward the government source of
life tables for men, I came up with the following table. I would like to
know if it makes sense.

Column 1 is your present age
Column 2 is your life expectancy at that age
Column 3 is the age that you have about a 50% chance of living to

45 32 77.5
50 27 78
55 23 79
60 19 79.5
65 16 80.5
70 13 82
75 10 84
80 7 86.5

Something about this table bothers me. Is the table approximately right? I
don't need it down the last decimal point. Does it basically add up this
way? I just want to be sure that I am not misinterpreting. To get the last
column I took the number of people left alive at that age and divided by 2.
Then I searched for that population count and saw at what age the life table
gave that population as still being alive.
Harrison

[Daniel J. Nordlund]

Harrison,

I don't know where your numbers are from, or whether you are interpreting them
correctly (or whether I am). But, if at age 45 you can "expect" to live
another 32 years, isn't what you want for column 3 simply the sum of columns 1
and 2?

Dan Nordlund

[Thomas Gatliffe]

 

Daniel J. Nordlund wrote:

>(snip)

>Column 1 is your present age
>Column 2 is your life expectancy at that age
>Column 3 is the age that you have about a 50% chance of living to
>
>45    32    77.5
>50    27    78
>55    23    79
>60    19    79.5
>65    16    80.5
>70    13    82
>75    10    84
>80    7      86.5
>
>Something about this table bothers me. Is the table approximately right? I
>don't need it down the last decimal point. Does it basically add up this
>way? I just want to be sure that I am not misinterpreting. To get the last
>column I took the number of people left alive at that age and divided by 2.
>Then I searched for that population count and saw at what age the life table
>gave that population as still being alive.
>Harrison
>

Harrison,

I don't know where your numbers are from, or whether you are interpreting them
correctly (or whether I am).  But, if at age 45 you can "expect" to live
another 32 years, isn't what you want for column 3 simply the sum of columns 1
and 2?

Dan Nordlund

Don't get mean and median confused.  Dan's answer approximates the mean age of death for a person currently 45.  Column 3 of your table approximates the median age of death for a person currently the age shown in column 1.  For younger ages the time to death variable is slightly left skewed and as the current age approaches the older ages the time to death becomes moderately right skewed.  The mean is less than the median in left-skewed distributions and the mean is greater than the median in right skewed distributions.

Bottom line:  Yes, your table does make sense.

Regards,

Tom Gatliffe

[Owner]

Thomas,

Thank you. It thought it was something like that. By the way I got the table from pulling data out of:

http://www.cdc.gov/nchs/data/sr2_129.pdf

"Thomas Gatliffe" <tho...@earthlink.net> wrote in message news:3951575F...@earthlink.net...