posted
There's thirty people at a party. How likely is it that two of those people have the same birthday? High? Low? So-so?
Posts: 722 | Registered: Jul 2004
| IP: Logged |
posted
What's interesting is that certain birthdays are more likely than others. Like nine months after winter.
Posts: 2705 | Registered: Sep 2006
| IP: Logged |
posted
In that case, I would love to do the math. Unfortunately, I just finished Stats, and am never doing any math again. Ever. Ever. Ever ever ever. I don't care if I accidentally pay 40k too much in taxes, as long as I don't have to do math.
Posts: 2705 | Registered: Sep 2006
| IP: Logged |
posted
I did this recently, and I forget the actual math and numbers and everything, but even with there only being 30 people, it is highly likely.
Posts: 349 | Registered: Oct 2007
| IP: Logged |
quote:Originally posted by SoaPiNuReYe: Somalian, are you taking Statistics by any chance?
Heh...no. That would be a clever way to get people to do your statistics homework problems for you though. Pose it as a challenge on internet boards! =D
Posts: 722 | Registered: Jul 2004
| IP: Logged |
posted
I posted enough information to definitively answer whether or not the chances would be high, low or so-so.
Posts: 722 | Registered: Jul 2004
| IP: Logged |
posted
I've heard this before. I think you do it by figuring out the probability of none of the people having the same birthday and then subtracting that number from 1.
If there were only two people in the group, the probability of them not having the same birthday would be the probability of person #2 not having the birthday of person #1 given the birthday of person #1, or 364/365. (We'll ignore leap years and assume the birthdays are evenly distributed.)
If there were only three people in the group, the probability of none of them sharing a birthday would be the probability of person #3 not having the birthday of either person #1 or person #2 given that persons #1 and #2 have different birthdays (which is 363/365) times the probability of persons #1 and #2 having different birthdays (which is 364/365 from above), or (363/365)(364/365). It's easier to see the pattern if you write it as (364/365)(363/365).
Continuing in this manner, we see that the probability that none of 30 people have the same birthday is the 29-term product (364/365)(363/365)...(336/365) = approximately .294, so the probability of at least 2 people out of 30 having the same birthday is approximately 1 - .294 = .706 = 70.6% (!). With leap years, the probability would be a little lower. I don't know offhand how to deal with them, though.
Posts: 781 | Registered: Apr 2005
| IP: Logged |
posted
Yup. I was thinking about what the chances were of two people having a specific date in common.
Posts: 3275 | Registered: May 2007
| IP: Logged |
quote:Originally posted by Omega M.: With leap years, the probability would be a little lower. I don't know offhand how to deal with them, though.
Here's my thought on that. First, though, let's increase the significant digits on your approximation, to 70.632%.
It seems to me that if we assume that one of the thirty people was born on February 29th, then the probability of at least one shared birthday, following the logic you laid out, is:
Now <insert much hand-waving here> the probability of any person having a birthday of February 29th (again assuming that birthdays are evenly distributed) is 1/1461. So I'm thinking that the overall probability of a shared birthday is:
quote:Originally posted by Icarus: Now <insert much hand-waving here> the probability of any person having a birthday of February 29th (again assuming that birthdays are evenly distributed) is 1/1461. So I'm thinking that the overall probability of a shared birthday is:
posted
What I'm interested in is the probability of running into someone with your exact birthday (month, day, and year), and who was born in the same hospital you were (so that you were babies in the nursery together), almost 40 years later and not in the general region where that birthplace is located (say, in the same state, but around 200 miles away).
Because that happened to me once, and it really freaked me out.
Posts: 2454 | Registered: Jan 2003
| IP: Logged |
quote:Originally posted by littlemissattitude: What I'm interested in is the probability of running into someone with your exact birthday (month, day, and year), and who was born in the same hospital you were (so that you were babies in the nursery together), almost 40 years later and not in the general region where that birthplace is located (say, in the same state, but around 200 miles away).
Because that happened to me once, and it really freaked me out.
Wait -- what are your parents's names? Oh no! Me too! But my parents put me and my sister up for adoption, because I was one of a set of triplets, and they didn't want to have more than one baby."
OK, not really, but it would be a hoot.
Posts: 10397 | Registered: Jun 2005
| IP: Logged |
quote:Originally posted by Tante Shvester: Yeah, I play that trick all the time.
"That's you're birthday? Mine too!
What year? Me too!
Where were you born? Really? Me too!
What hospital? Me too!
Wait -- what are your parents's names? Oh no! Me too! But my parents put me and my sister up for adoption, because I was one of a set of triplets, and they didn't want to have more than one baby."
OK, not really, but it would be a hoot.
Point taken.
However...I saw the driver's license (she came through my checkout line when I was working in retail, and she had to show me her DL because she paid by check), so the birthdate was legit, and she was the one who stated her birthplace and I know where I was born and when, and they matched. So, in this case it couldn't have been that old trick.
Posts: 2454 | Registered: Jan 2003
| IP: Logged |
Ahem. February 29th is the 366th day. Non-leap years have 365 days.
quote:Wait -- what are your parents's names? Oh no! Me too! But my parents put me and my sister up for adoption, because I was one of a set of triplets, and they didn't want to have more than one baby."
posted
There are about 15 people in the newsroom where I work, and I share a birthday with one of them. We share a few personality traits, as well.
Posts: 2034 | Registered: Apr 2004
| IP: Logged |
posted
And because birthdays are not randomly distributed. As someone upthread mentioned, September babies are pretty common.
Posts: 5462 | Registered: Apr 2005
| IP: Logged |
posted
The ethics of that study (and the doctor whose brainchild it was) have been debated pretty hotly for quite some time.
Posts: 32919 | Registered: Mar 2003
| IP: Logged |
posted
That would be a weird study to inherit, as a researcher. The separation is one thing, but it also seems like they would have to be keeping tabs on these people. And what kind of a study drops people like this? Did they have to manifest mental illness by a certain age or something? Stay living in New York? It's just super weird.
Posts: 11017 | Registered: Apr 2003
| IP: Logged |
quote:Originally posted by littlemissattitude: What I'm interested in is the probability of running into someone with your exact birthday (month, day, and year), and who was born in the same hospital you were (so that you were babies in the nursery together), almost 40 years later and not in the general region where that birthplace is located (say, in the same state, but around 200 miles away).
The answer is 0. I was born in a small, rural hospital. I was the only baby born there on that day (actually, I think I was the only baby born there that month)!
Posts: 340 | Registered: Mar 2003
| IP: Logged |