Talk:Chance News 24

From ChanceWiki
Revision as of 11:53, 20 March 2007 by Shoup (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

[I hope this is the appropriate place for this comment.] I always enjoy the Chance News, but the story on the PEAR Lab (CN#24) is really not up to its usual high quality level.

Just one example is the sentence "The beginning is meager because there is a suspicion that random number generators aren't all that accurate." First of all, RNGs are not judged on some sort of "accuracy", but perhaps on producing uniform distributions. Secondly, if one actually takes the trouble to *read* the PEAR Lab reports before lampooning them, you will find careful calibration runs throughout, and all experiments done trivalently, with aim-high, aim-low, and no-attention runs for each subject, etc.

And Robert Park is really not an appropriate critic concerning this sort of work, since he too apparently hasn't really looked at or understood it, but just argues by appeal to ridicule -- the worst sort of pseudo-skepticism. There are serious discussions to be had surrounding all leading-edge research of this kind, but Park's writings are not the place to look.

A healthy skeptical attitude is an important part of statistical studies and investigations, but this CN article shows just the opposite, and I for one am disappointed.

Richard Shoup, Boundary Institute, shoup@boundary.org

--Shoup 11:53, 20 Mar 2007 (EDT)

Personal tools
Namespaces
Variants
Actions
Navigation
Toolbox