[extropy-chat] the structure of randomness

gts gts_2000 at yahoo.com
Fri Dec 30 19:02:22 UTC 2005


On Fri, 30 Dec 2005 12:58:32 -0500, Damien Broderick  
<thespike at satx.rr.com> wrote:

> http://www.21stcenturysciencetech.com/articles/time.html
>
> It appears to show some surprising regularities within or atop various  
> stochastic structures, correlating with, for example, sidereal rather  
> than terrestrial/solar time. A possible connection between global and  
> quantal scales?

Standard statistical tests exist for determining if a given distribution  
is normal. I don't see in this article any reference to such testing,  
which is not to say they did none. From reading the article alone it seems  
they may have relied on a purely visual analysis of the distributions, and  
then "devised a computer-based algorithm for measuring the relative degree  
of 'closeness' or similarity of histogram shapes, and on this basis  
carried out a computer analysis of hundreds of histograms taken over a  
long period."

I'm not sure their home-brewed algorithm would have any value if the  
compared distributions were not truly abnormal in the first place. Looks  
can be deceiving, and the distributions do look roughly bell-shaped.

-gts




More information about the extropy-chat mailing list