I'm giving up on PGP

Re:I'm giving up on PGP by jballanc on Tuesday December 06, 2016 @06:54PM (#13119331)

I think a Poison distribution would be more appropriate here, so the probability of 1 man in 5 being colorblind, assuming 10% of the population on average is colorblind, is:

    e^-0.5*sum((i)->(0.5^i/factorial(i)), 1:5) = ~39%

[ Parent | on HN ]

Re:I'm giving up on PGP by softawre on Tuesday December 06, 2016 @07:48PM (#13119610)

> Poison distribution

Why? This is basic statistics (% of population).

[ Parent | on HN ]

Re:I'm giving up on PGP by caf on Tuesday December 06, 2016 @10:17PM (#13120181)

I don't believe so - it's n instances of a bi-valued random variable, IID. It's exactly the case the binomial distribution covers.

[ Parent | on HN ]

Re:I'm giving up on PGP by mnw21cam on Wednesday December 07, 2016 @04:21AM (#13121409)

Exactly. You have five chances at an event, and an event probability per chance, which is exactly what the binomial distribution is for.

The Poisson distribution is what you use when the effective number of chances is "large". The Poisson distribution is effectively a special case of the binomial distribution, where the number of chances is infinity, the probability per chance is infinitesimal, and the product of the two is the expected number of events.

[ Parent | on HN ]