Daft probability question

[Julia]

Hello, my name's Julia and I'm a mathematical duffer. Could one of you clever people therefore answer the following question:

What are the odds of a person being born and dying on the same calendar date? I would have thought 365 to 1 but it sounds too bleedin' obvious.

[brianp]

What are the odds of a person being born and dying on the same calendar date? I would have thought 365 to 1 but it sounds too bleedin' obvious.

Bleedin' obvious but pretty much spot on. It may be very slightly lower than 1:365 because I seem to recall (though I can't find a reference) that one's birthday, along with Christmas & New Years Days, are popular days to commit suicide.

I'm actually on for a Triple - I got married on my birthday, so maybe it will be a record if I die on that day too.

[fotworth]

Hello, my name's Julia and I'm a mathematical duffer. Could one of you clever people therefore answer the following question:

What are the odds of a person being born and dying on the same calendar date? I would have thought 365 to 1 but it sounds too bleedin' obvious.

there are 365 days in a year so that would make the odds 364/1 if all years was the same.
A leap year has 366 days making the odds slightly longer. If you was born on the 29th of feb it gets realy diffecult to count.
So we are both wrong but Im more right than you

[brianp]

there are 365 days in a year so that would make the odds 364/1 if all years was the same.
A leap year has 366 days making the odds slightly longer. If you was born on the 29th of feb it gets realy diffecult to count.
So we are both wrong but Im more right than you

I wasn't wrong - I said 1:365 was "pretty much" spot on, not precisely correct.

Where the blazes do you get 1:364 ? If we assume that death can occur on any day in the year with equal likelihood then odds of 1:365, 1:366 and even 1:365.25 are defensible, never 1:364.

If you die in a non-leap year the odds of dying on your birthday are 1:365 and the odds of dying on a day other than your birthday are 364:365 - and the two add to give 365:365, ie certainty, because you have to do one or the other.

If you were born on 29 Feb the odds of dying on your birthday in a given year are either zero (if that year doesn't have a Feb 29) or 1:366 (if it does.) Taken over a 4 year period you have 1461 days and one Feb 29 - so over that period the odds of dying on your birthday are 1:1461.

[Pebble]

I think this is just another version of the birthday question:

http://en.wikipedia.org/wiki/Birthday_problem

In probability theory, the birthday problem, or birthday paradox[1] pertains to the probability that in a set of randomly chosen people some pair of them will have the same birthday. In a group of at least 23 randomly chosen people, there is more than 50% probability that some pair of them will have the same birthday. Such a result is counter-intuitive to many.
For 57 or more people, the probability is more than 99%, and it reaches 100% when, ignoring leap-years, the number of people reaches 366 (by the pigeonhole principle). The mathematics behind this problem led to a well-known cryptographic attack called the birthday attack.

The only difference is that where the perinatal mortality is high, the liklihood of dying on the exact same day as you are born increases.

[brianp]

I think this is just another version of the birthday question:

http://en.wikipedia.org/wiki/Birthday_problem

No, it's completely different. In the birthday question we have a group of people and we are asking for the probability the any two of those people will share a birthday. So, to give a specific example, let's say we have 25 people. Each will have a birthday giving us 25 dates - and we want the probability that two or more of those 25 dates are identical. It's a very complicated question with a very non-intuitive result.

In Julia's scenario we have a single person with a single birthday and we want to know the probably that a random event - that person's death - will occur on that same date. A very simple problem with a very simple and intuitive result - approximately 1:365.

[Pebble]

No, it's completely different. In the birthday question we have a group of people and we are asking for the probability the any two of those people will share a birthday. So, to give a specific example, let's say we have 25 people. Each will have a birthday giving us 25 dates - and we want the probability that two or more of those 25 dates are identical. It's a very complicated question with a very non-intuitive result.

In Julia's scenario we have a single person with a single birthday and we want to know the probably that a random event - that person's death - will occur on that same date. A very simple problem with a very simple and intuitive result - approximately 1:365.

I am not sure - you are right it is not the same as birthdays, because these are two completely independent events.

We have two events - one birth one death. For any one individual you are correct, the birthdate is already decided, therefore that individuals death can only fall on one day in the year (365/365)x(1/365). However, if the question is what are the chances of some one being born & dying on the same day then we could choose a sample of say 60,000,000 - and their birthdates will be randomly assorted over the 365 days of the year, their dates of death will also be randomly assorted over the same 365 days of the year. So you do not need to review 180 deaths to have greater than a 50% probability of finding someone with the same date of death and birth. I'll bet Matt knows the answer.

[Pebble]

I am not sure - you are right it is not the same as birthdays, because these are two completely independent events.

Brianp: The more I think about this, the more I think you are right. Given that one is dealing with conditional probability, intuitively 1/365 goes against the grain with me. However, the first event does not in any way alter the probability of the second event.

[FarSideOfTheMoon]

I think it is as simple as it seems, if you keep it simple.

If you did an analysis of a large set of historical data you may find for out that some people die before their first birthday, not sure if that would have an impact?

If you say what is the probability of me dying on my birthday in the next year - that may depend on which date you make the prediction. If you are aged 80 and your birthday is 10 or 11 months away, maybe there is a greater chance you die before you reach your birthday. Again, would that affect the odds even to a very small degree?

If you are making a general statement without worrying about such complexities, then it is a simple 365-1 or 366-1, or if you want to include say 10 yearsr prediction, then it would be something like 365.2 to 1, to take account of some years are longer than others.

[fotworth]

I wasn't wrong - I said 1:365 was "pretty much" spot on, not precisely correct.

Where the blazes do you get 1:364 ?.

If you have 6 dogs in a race and trap 6 wins there was only 5 other possable winners. Traps 1 2 3 4 & 5
That would give you odds 5 against 1 expressed as 5/1.
Therefor if you look at a 365 day year. and only 1 day matches the birth date,That leaves 364days that dont match.
Thats 364 possable days to die that dont match the birth day against the 1 day that matches the birth day. expressed as 364/1
How the hell do you work your odds out. Give me the odds of a dog winning a six dog race.

If you were born on 29 Feb the odds of dying on your birthday in a given year are either zero (if that yeaIr doesn't have a Feb 29) or 1:366 (if it does.) Taken over a 4 year period you have 1461 days and one Feb 29 - so over that period the odds of dying on your birthday are 1:1461.

Your math is quite terrable in a 4 year period you have 1461 days including Feb 29. Not and one Feb 29. How can I accept your answer if you cant express the equation
If you died at the age of 5 or the age of 105 throws your reconing futher into the realms of guess.
Someone born on feb 29th and dies at age 5 saw only 1 other feb 29th within the life time.
where as Someone born on feb 29th and dies at age 105 sees 41 feb 29ths within the life time.
So the elder has forty more chances than the child 40/1 against the child.

I wasn't wrong - I said 1:365 was "pretty much" spot on, not precisely correct.

There is NO finite answer.
Give me the odds of a dog winning a six dog race. Or simpler still 2 dog race.

[brianp]

If you have 6 dogs in a race and trap 6 wins there was only 5 other possable winners. Traps 1 2 3 4 & 5
That would give you odds 5 against 1 expressed as 5/1.
Therefor if you look at a 365 day year. and only 1 day matches the birth date,That leaves 364days that dont match.
Thats 364 possable days to die that dont match the birth day against the 1 day that matches the birth day. expressed as 364/1
How the hell do you work your odds out. Give me the odds of a dog winning a six dog race.

Your math is quite terrable in a 4 year period you have 1461 days including Feb 29. Not and one Feb 29. How can I accept your answer if you cant express the equation
If you died at the age of 5 or the age of 105 throws your reconing futher into the realms of guess.
Someone born on feb 29th and dies at age 5 saw only 1 other feb 29th within the life time.
where as Someone born on feb 29th and dies at age 105 sees 41 feb 29ths within the life time.
So the elder has forty more chances than the child 40/1 against the child.
There is NO finite answer.
Give me the odds of a dog winning a six dog race. Or simpler still 2 dog race.

Well you've certainly demonstrated that you don't know the first thing about probability, and as you've amply demonstrated on other threads that you're an idiot too, I've no intention of wasting my time trying to correct your erroneous ideas.

[fotworth]

Well you've certainly demonstrated that you don't know the first thing about probability, and as you've amply demonstrated on other threads that you're an idiot too, I've no intention of wasting my time trying to correct your erroneous ideas.

LOL If you want to try and prove it I'll give you 1/1, even money Heads I win tails you loose

[Croydon Bob]

Your math is quite terrable

Your maths is as "terrable" as your spelling dumbo. There is no reason to remove the birthday, it isn't eliminated as an option unlike in your dog analogy. You are wrong, Brianp is correct. But no doubt you are too stupid to see it and will whine on pointlessly digging yourself ever deeper.

[fotworth]

Your maths is as "terrable" as your spelling dumbo. There is no reason to remove the birthday, it isn't eliminated as an option unlike in your dog analogy. You are wrong, Brianp is correct. But no doubt you are too stupid to see it and will whine on pointlessly digging yourself ever deeper.

I am more right than all of you cuz youre THICK.
The calinder date for today is the 5th of march 2010 the odds of a person being born and dying on the same calindar date 5th of march 2010 can not be calculated. And if it could not all the calendar dates would give the same results. Do you realy think 1:36 is the pretty much" spot on, not precisely correct answer? I'll cut and paste the question for you to read again .

What are the odds of a person being born and dying on the same calendar date?
Answer DAFT question.

nice one julia

[Pebble]

I am more right than all of you cuz youre THICK.
The calinder date for today is the 5th of march 2010 the odds of a person being born and dying on the same calindar date 5th of march 2010 can not be calculated. And if it could not all the calendar dates would give the same results. Do you realy think 1:36 is the pretty much" spot on, not precisely correct answer? I'll cut and paste the question for you to read again .

What are the odds of a person being born and dying on the same calendar date?
Answer DAFT question.

nice one julia

I fail to follow your logic and I have given this a little thought. Certainly if you die in the first year of life then you are likely to have been so sick at birth that there is a slightly greater chance of it not being your birthday - however with an average survival of close to 80 years this can be discounted. How long you live beyond this point is irrelevant - since you will die, and it will be on a day of the year, therefore for each year that passes (forget about 29th Feb doesn't contribute significantly) you have 364 chances of not dying on your birthday and 1 chance of dying on your birthday. The same is true of everyone else. The date of your birth has no significant impact (accepting that some dates are associated with more suicides), so the probability of dying on any particular date is not predetermined in any way by selecting date of birth.
So much as the conlcusion irks me, I have to accept that 1/365 is right (approx).