Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Standard deviation calculation.

1 view
Skip to first unread message

Octo the Genarian

unread,
Sep 28, 2003, 1:18:55 AM9/28/03
to
In another thread Bill Chen calculated the variance of the number of hands
won by a given player in a 6-man field over 78 hands as:

78 * 1/6 * 5/6 = 10.83

And then of course the SD is the square root of that.

My question is how did he arrive at that original calculation? I only know
how to calculate SD from a sample.


Mike

unread,
Sep 28, 2003, 1:37:37 AM9/28/03
to

It's only been about a yr since I last took a stats class, and I've
forgotten much of the material, but I'm pretty sure this is a binomial
distribution, and Bill calculated it as such.

Regards,
Mike

_________________________________________________________________
Posted using RecPoker.com - http://www.recpoker.com


Octo the Genarian

unread,
Sep 28, 2003, 1:57:31 AM9/28/03
to

Thanks.

If anyone else is interested, see
http://www.stats.gla.ac.uk/steps/glossary/probability_distributions.html#binodistn.

Forget shopping, math is easy!

"Mike" <joec...@hotmail.com> wrote in message
news:3f7673a0$0$210$9a6e...@news.newshosting.com...

Harry Clyde

unread,
Sep 28, 2003, 3:51:27 AM9/28/03
to

If X~Binomial(n,p) then V(X) = np(1-p)

where n is the number of trials, p is the probability of a success


Garycarson1

unread,
Sep 28, 2003, 6:03:43 AM9/28/03
to
In general, V(X) = E(X^2) - E(X)^2

The binomial formula for variance is just a special case of that, when there
are only two possible outcomes.

When there are 3 possible outcomes in your model (say you might allow for
winning a big pot, or lose a big pot, or lose a small pot are the three
possible outcomes) then use the general equation.

Gary Carson

0 new messages