[CoLoCo] dice rolls in C

[Richard Guenther]

I believe you'd use a T-test for comparing the means of those two tests (by
hand vs. RNG).  If you did dice rolls and counted the numbers of 1s, 2s, and
so on, you could do a chi-squared analysis.  Those tests would only tell you
so much, though.  If you could compare them to a normal distribution, that
might be better, as it would take into account the standard deviation.

Here's a good site:

http://www.graphpad.com/www/book/Choose.htm

I would think any test will show no significant difference, although I'm not
sure if we really know exactly what test we'd like to show that there is a
difference between real dice and RNGs.

Have fun,

Richard


On Feb 10, 2008 10:55 AM, Jim Hutchinson <jim at ubuntu-rocks.org> wrote:

>
> What I'd really like to figure it is what stats to use to compare 1000
> real dice rolls to 1000 computer dice rolls and see if the differences
> are statistically insignificant.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://lists.ubuntu.com/archives/ubuntu-us-co/attachments/20080210/b8d82c59/attachment.htm