Life expectancy as a probability distribution curve

[clop]

I'm been searching the net for a while now but I can't find what I'm looking for. I'm not even sure that what I'm looking for even exists. Is there such a thing as a graph showing life expectancy as a probability distribution? You know the kind of thing, with age along the x-axis and the probability of dying at that age on the y-axis, with the total area beneath the curve equal to 1. I'd be interested to see how these curves have changed since mortality records began, and also the difference between the curves by sex and by country.

clop

[Forskern]

US statistics for 2003 in this 40 page PDF:
http://www.cdc.gov/nchs/data/nvsr/nvsr54/nvsr54_14.pdf

No plots, but the data you wanted is tabulated, and I guess US 2003 isn't to different from 2006 Oz

[clop]

Thank you

[peteshimmon]

This is something I have wanted to see for a
while despite it seeming a bit morbid. However a
British publication from the National Statistics
office has some graphs. It is remarkable how the
annual number of deaths throughout the 20th
century hovered around 600,000 with the
population increasing from 40 million I think.
I would like a short filmstrip of each years
mortality in histograms of one year bins. The
hump of old age deaths advancing in years over
this time should look quite dramatic. It has
implications for peoples reaction to pension
exhortations from governments. They want folk to
save more but people can have pesimistic ideas
and spend while they can. Overall the saving
might be optimum!

[JohnD]

clop,
Have you used the words "actuary" or "actuarial" in your search?
An actuary is one who calculates insurance risks, especially for life, or retirement insurance, so they know quite a lot about this!
Does this site help? http://www.michigan.gov/orsschools/0...486---,00.html

John

[grant hutchison]

JohnD's suggestion is good. There's a type of graph very similar to what you describe, clop, called an "actuarial curve". It plots age against your probability of being alive at that age.
A search strategy based on just the phrase "actuarial curve" isn't very helpful though, since the same graphing technique can obviously be used for small subsets of the population, or for outcomes different from "being alive": being free from cancer recurrence or free from repeat surgery, for instance. So you'd need to refine the search with other terms - but it might help in your quest.

Grant Hutchison

[mike alexander]

A while back I took a bunch of issues of my alumni magazine and tallied up the deaths by class year. With about a thousand names it very nicely produced a standard mortality curve. In one of those odd involutions of life, I wrote an essay about it and the same magazine bought it.

[JohnD]

Thnaks for that addition, Grant.
Clop, how about "life expectancy"? This lead me to the UK Government Actuary Department website (God's Teeth! There really is everything on the InterNet!) at http://www.gad.gov.uk/Life_Tables/In...ife_tables.htm

I think it includes exactly what you want - "Interim Life Tables"

Bon appetit!

John