Re: statistics silly question??
- From: hrubin@xxxxxxxxxxxxxxxxxxxx (Herman Rubin)
- Date: 2 Mar 2006 12:47:19 -0500
In article <3904466.1141173905687.JavaMail.jakarta@xxxxxxxxxxxxxxxxxxxxxx>,
tommy1729 <tommy1729@xxxxxxxxxxx> wrote:
hi everyone
here comes a possible silly question.
I like my other 2 posts better but i just had to ask :p
Gauss "transformed" the binomial distribution into the erf(x)
This is false.
i don't know how , i suppose he used limits and integrals.
at least that's how i can explain it, so seems likely to me.
It was done by de Moivre, long before Gauss was born.
By using the leading term of Stirling's approximation to
the factorial, the binomial probability is close to the
density of a normal distribution with the same mean and
variance. Just as the integral is approximated by the
sum, so is the sum by the integral; see the Euler-Maclaurin
summation formula.
is there another way ?
There are many, none so enlightening.
and how about a generalization kind a like this:
"transform" the trinomial distribution ?
to understand trinomial distribution its like the gausscurve ( discreet) of the trinomial coefficients , there like binomium but instead of (1+x)^n ; it's (1+x+x^2)^n .
No problem; use the same approximation to the factorial.
It will work for arbitrary multinomial, and even for sampling
with an infinite number of choices.
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
hrubin@xxxxxxxxxxxxxxx Phone: (765)494-6054 FAX: (765)494-0558
.
- Follow-Ups:
- Re: statistics silly question??
- From: tommy1729
- Re: statistics silly question??
- References:
- statistics silly question??
- From: tommy1729
- statistics silly question??
- Prev by Date: Re: Example to show a limitation of the Runge-Kutta Method
- Next by Date: Re: Beal's conjecture
- Previous by thread: Re: statistics silly question??
- Next by thread: Re: statistics silly question??
- Index(es):
Relevant Pages
|