[Date Index] [Thread Index] [Author Index]

Re: bimodal ditribution form counting signs of Pi digits differences

• To: mathgroup at smc.vnet.net
• Subject: [mg51857] Re: bimodal ditribution form counting signs of Pi digits differences
• From: Bill Rowe <readnewsciv at earthlink.net>
• Date: Wed, 3 Nov 2004 01:26:27 -0500 (EST)
• Sender: owner-wri-mathgroup at wolfram.com

On 11/2/04 at 10:29 AM, drbob at bigfoot.com (DrBob) wrote:

>Why isn't the KS table built into Mathematica, I wonder? Or is it,
>somewhere?

As far as I know, the KS table is not built in to Mathematica nor is it a function in any of the packages that come in the standard distribution.

FWIW,

1/2Log[1/(1-p)] - 1/(6 Sqrt[n])

is a good approximation to the pth percentile point for the KS statistic with an error O[1/n] for n>30
--
To reply via email subtract one hundred and four