At 12:02 PM 5/21/2012, Jed Rothwell wrote:
Here are age distributions for England and Wales, 1696:
Age group, percent
0 - 9 27.6%
10 - 19 20.2%
20 - 29 15.5%
30 - 39 11.7%
40 - 49 8.4%
50 - 59 5.8%
60 and above 10.7%
Laslet, p. 103. Those numbers are reliable. They kept good records in the U.K.
So, 11% lived to their 60s, and there were more elderly people over
60 than people in their 40s or 50s. The average age was 27.5.
One-third to half of burials recorded in a French 17th century
record were listed as "children," meaning they were probably under
20 -- too young to be listed by employment. If you reached age 20
and you were still healthy, and you had acquired immunity to measles
and smallpox, you had a good chance of seeing age 60. You did not
drop dead 7.5 years later.
I can't find specific papers or data, but I suspect that the human
population is not homogenous, consisting of several overlapping
species (male, female for starters), and that the "centenial" cohort
simply does not die off from disease. Their peak on a normal
distribution might be at 60, so they would hardly show up for deaths
at earlier ages.
googling : human longevity long tail (in the statistical sense!)
indicates that a fair number of people are looking for genetic
markers to explain it.
And of course, the people most WORRIED about it are actuaries:
http://www.partnerre.com/reviews/article/stochastic-model-longevity-risk
disclaimer : my usual speculative, ignorant rambling opinions, of course ...
