[R] Size of observations

Charles C. Berry cberry at tajo.ucsd.edu
Tue May 16 18:29:33 CEST 2006


This isn't an R question is it?

 	PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html


But I know why this happened, so I will answer your question anyway.

During the first few months of rolling your die, it really was not 
uniform and this is what accounts for the results after a mere 8,000,000 
rolls.

But after the first year or two of continuous rolling the corners wear 
off the die - it looks more like a sphere --- check what yours looks like 
- it really isn't a cube like shape anymore is it??

When that happens the distribution of results looks more uniform. But you 
probably cannot still read the little dots on the sides anyway - the last 
time I did this myself they wore off after about 14,275,839 rolls.

HTH!

On Tue, 16 May 2006, Amit Kabiri wrote:

> Hi All
>
>
>
> I had to assure the randomness of backgammon game's two dices.
>
>
>
> I took 8,000,000 observations and found that one of the dices is not uniform
> distributed.
>
>
>
> I took 80,000,000 observations and found that the dice is uniform
> distributed.
>
>
>
> In both tests I used Pearson Chi Square test.
>
>
>
> My questions:
>
>
>
> 1.	How many observations should I take?
> 2.	How much the 80,000,000 observations test is batter from the
> 8,000,000 test?
>
>
>
>
>
> Thanks in advance for your help
>
>
>
> Amit Kabiri
>
> akabiri at 012.net.il
>
>
>
>
>
>
> 	[[alternative HTML version deleted]]
>
>
>
>
>    [ Part 3.58: "Included Message" ]
>

Charles C. Berry                        (858) 534-2098
                                          Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	         UC San Diego
http://biostat.ucsd.edu/~cberry/         La Jolla, San Diego 92093-0717




More information about the R-help mailing list