Normally DeAd?

[LOST6200]

It seems sliek a lots of dead people in the groupd. How do I do the acturearials caluctions for normals? For exmpale: if there are 100 people born in 1959 that are stil alive in 1972, waht percentage should statistically be aliuve todya?

 

[Pradeep]

I guess you would find the average life expectancy for someone born in 1959, and go from that. To be more accurate you would need to know the male/female population of the 100, as females seem to live longer on average. There's probably some distribution curve where you increase the chances of carking it as you get older.

 

[Pradeep]

I guess you would find the average life expectancy for someone born in 1959, and go from that. To be more accurate you would need to know the male/female population of the 100, as females seem to live longer on average. There's probably some distribution curve where you increase the chances of carking it as you get older.

Edited to add, live life like you may die tomorrow, and you won't have any regrets.

 

[jtr1962]

From what I understand about the way these actuarial tables work, they tell you statistically if you're a given age now, then you can expect to live a certain amount of years. There is no simple equation. It's just a table based on statistics. Interestingly, the average lifespan is NOT how long the average person can expect to live to since that average includes many who die young from accidents or hazardous hobbies/occupations. For example, I read that a male born today who takes reasonably good care of themselves (i.e. not severely overweight and sedentary) can expect to live to about 86 on average while a similar female can expect to live to 89. However, as we have medical advances plus a hopefully cleaner environment in the future these numbers may well increase so that by the time a male born today reaches age 86 he might expect to live another 20 or 30 years, on average. Genetics and lifestyle of course also play a large part. You can probably extend your expected lifespan by 20 years or more by taking exceptional care of yourself. It'll also mean that you'll enjoy greater mobility as you age which translates into improved quality of life. On the other hand, neglecting your health can take years off your life. I read that just smoking alone reduces lifespan by an average of 15 years.

 

[LOST6200]

OK, but wher cvan I get that ifo. Or is it something very expensive with a sucsrtiction or serbvice fee, only avaaailable gto the insurance industry?

 

[Tea]

I read that smoking alone reduces lifespan by an average of 15 years.

Yike! Lucky for me that I'm a party girl and only smoke when there are other peope around. Tannin is in trouble though, as he often goes out and sits by himself for a quiet smoke on the back steps. If he's on the porch and I'm inside using the computer, does it still count as smoking alone?

Hey! What about drinking alone? Is that safe?

 

[Tannin]

Stupid ape.

Interestingly, the average lifespan is NOT how long the average person can expect to live to since that average includes many who die young from accidents or hazardous hobbies/occupations.

In other words, it's not a proper average at all. There are several ways to calculate an average, and no competent or honest statistician would report the average of a skewed distribution using the arithmetic mean. For skewed distributions, you use the median. This is really basic stuff that you should learn in fifth grade. You can only report the mean in a valid way where the distribution is roughly even (i.e., not skewed to either end of the graph).

To calculate the real average lifspan, simply take the median. Say you have 100 people. The average lifespan is the age of the 50th person to die. Even an ape could work that one out.

 

[Tea]

meanie!

 

[mubs]

To calculate the real average lifspan, simply take the median. Say you have 100 people. The average lifespan is the age of the 50th person to die. Even an ape could work that one out.

I don't think so. I was in fifth grade a loooong time ago, but my recollection is thus: Median should be the mid point of the age at death of the 100 persons. In other words, the midpint (hic) between the youngest person who died in the group of 100, and the oldest person who died in the group of 100.

 

[Tannin]

That's exactly right, Mubs. And exactly what I said too. From Wikipedia: "the median of a finite list of numbers can be found by arranging all the observations from lowest value to highest value and picking the middle one".

To calculate average lifespan, first you determine that the distribution is skewed, thus the correct measure of centeral tendency (i.e., "average") is the median. Then you make a list of all the individual lifespans (i.e., ages at death), and finally you pick the middle one. There is your answer.

 

[ddrueding]

Now all we need is a list of relevant ages. For Eric; we need to know the year of your birth, and how many of the people born that year are still alive. If it's less than half...count yourself lucky and double-check that I'm in your will.

 

[Sol]

I think what mubs is suggesting is that you take the highest and lowest ages and add half the difference between them to the lowest age to determine the median. So technically no one would actually have to die at a specific age for it to be the median.

A quick search for definitions suggests that roughly 60-70% of sites define the median similarly to mubs and the rest go with Tannin... I seem to recall the median always being part of a the set of numbers but primary school was a long time ago even for me...

 

[ddrueding]

Sol, I can't think that that is right. Placing so much significance on a single (the lowest) value could potentially really screw up the average. IIRC, one of the reasons for choosing Median instead of Mean or Mode was to prevent outliers from having a significant impact on the calculation.

 

[Tannin]

Sol, that is the most ridiculous thing I ever heard. Wrong, wrong wrong. I don't even know what that weird thing you just described to me is called (and I have a degree in, among other things, statistics). So far as I know, it doesn't even have a name, because it doesn't exist.

It's not hard to figure out why when you think it through: that method discards the information from every score except the two that are least representative of central tendency. What a joke.

 

[Buck]

I'm no expert on median/mean, but, does this help?
http://en.wikipedia.org/wiki/Median

Popular explanation
The difference between the median and mean is illustrated in a simple example. Suppose 19 paupers and 1 billionaire are in a room. Everyone removes all money from their pockets and puts it on a table. Each pauper puts $5 on the table; the billionaire puts $1 billion (that is, $109) there. The total is then $1,000,000,095. If that money is divided equally among the 20 persons, each gets $50,000,004.75. That amount is the mean (or "average") amount of money that the 20 persons brought into the room. But the median amount is $5, since one may divide the group into two groups of 10 persons each, and say that everyone in the first group brought in no more than $5, and each person in the second group brought in no less than $5. In a sense, the median is the amount that the typical person brought in. By contrast, the mean (or "average") is not at all typical, since no one present—pauper or billionaire—brought in an amount approximating $50,000,004.75.

 

[LOST6200]

Wow, so manth not mth iwzzes. iknow how the caluclatiuon for the mean and could even do the probablitly. However I bneed to know teh distribution of daeth probablity at least with 1-year intervals. Whetr is be the data found to perforem the caluculations? Thanks.

 

[Tea]

Do you need that for the US, Lost One? Or could you make do with figures from a broadly similar country, with an approximately similar lifestlye? Oz wouldn't be too different, I should think. If Oz figures would do, you would almost certainly find what you need at the Australian Bureau of Statistics, there is a heap of stuff there. Address should be http://www.abs.gov.au

 

[fb]

I think WHO also has some statistics for most countries, but it's a few years old.

 

[mubs]

Tannin and I said the same thing, with one crucial difference: in his first post defining median, he forgot to mention that the 100 ages had to be sorted first... the age of the 50th person to die, going chronologically in an unsorted list, may or may not be the median.

He clarified this in response to my post.

Me at my nitpickin best. :-D

 

[ddrueding]

Mubs, I know that it's 6:45AM here, and that I am awful this early in the morning; my brain simply doesn't work. But if you are waiting for the 50th person to die, they have sorted themselves

 

[Tannin]

Sorry Mubs: I was assuming (from a post way up above somewhere) that we were starting with 100 people all born at around about the same time.

 

[Sol]

Wow jump down my throat there Tony, I said that's what I thought Mubs was saying and then proceeded to agree with you... So we were actually all in agreement the whole time...

 

[LOST6200]

Now I'm sorryy for the asking... 21% personnel deceeased just seesm rather high, and why not me amognst the dead ones?

 

[CougTek]

You'll die in 11 years, 106 days, 5 hours, 54 minutes and 42 seconds at the moment this message is posted. Now you know, happy now?

 

[ddrueding]

21% deceased isn't that high a percentage, really. Tony might back me up on this (or jump down my throat ) but this kind of thing has significant outliers (think asymmetrical bell curve). I'm much younger than you, and I would bet that it's already in the double digits percentage wise.

 

[time]

Sol, all that living amongst British people has warped your sense of reality. Next thing we know, you'll be telling us that England has a cricket team and they can win matches! Get a grip, Man.

 

[Sol]

Actually I was in Italy visiting my sister for Christmas whilst the ashes were on. I haven't heard the word cricket spoken once since getting back... Not only don't they have a team that can win but I'm not even sure they'd recognize the name of the game these days...