[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math

View post   
File: 45 KB, 313x192, We should be friends.png [View same] [iqdb] [saucenao] [google]
7096821 No.7096821 [Reply] [Original]

Hey /sci/, first time poster in this board. I have a doubt about probability which is nagging at my mind,and I hope you're gonna help me solve this.

Let's say I pick a completely random natural number x. There are no limits on how big x can be, it can be any number in N. If A means x is an even number and B means x is divisible by 13, is P(A) > P(B)?

>> No.7096825

>Let's say I pick a completely random natural number x

meaningless statement unless you define the probability distribution

>> No.7096834

>>7096825
Is it enough if I say every number has the same probability and the sum of the probabilities of all numbers equals 1?

>> No.7096845

>>7096834
This is ill posed. There is no such finite number.

>> No.7096853

>>7096845
I'm sorry, I'm not really an expert, just curious. Could you explain why it's not possible?

>> No.7096871

>>7096825
When normal people say that they mean uniformly distributed

>> No.7096885

>>7096853
If you pick any positive number for the probability, the sum will diverge.

>> No.7096903

>>7096885
Understood it now. Well, thank you, anon. "Your question is flawed" is not the most satisfying of answers, but it's a good answer nonetheless.

>> No.7096923

>>7096871
There is no uniform distribution on an unbounded infinite set, you fucking retarded piece of shit.

>> No.7097705
File: 23 KB, 500x465, mmmmmmhmmmmmm.jpg [View same] [iqdb] [saucenao] [google]
7097705

>>7096871

>uniformly distributed
>between zero and infinity

>> No.7097844

>>7096871
>normal people
people in a normal distribution?

>> No.7097877

>>7096923
>>7097705
for you assburgers ITT, let the distribution be 1/n for each integer from 0 to n, then take the limit of n -> infinity.
its then easy to show that P(A) > P(B) in the limit.

>> No.7097897

>>7097877
That's not a probability distribution. The probabilities don't sum to 1.

>> No.7097908

>>7097897
1/n*n ist 1?
il be damned

>> No.7097921

>>7097908
it's probably because you said form 0 to n which would be (1/n) * (n+1).
It would be correct from 0 to n-1 or form 1 to n

>> No.7097949

>Is it enough if I say every number has the same probability
Ignoring all the autism about definitions
P(A)= 1/2, P(B)= 1/13
therefore P(A) > P(B)

>> No.7097962

>>7097921
n+1 or n doesn't matter in the limit n -> infinity.