Author: Rolf Tueschen
Date: 18:25:58 12/30/05
Go up one level in this thread
On December 30, 2005 at 15:06:34, Ingo Althofer wrote: >On December 30, 2005 at 13:47:15, Roger D Davis wrote: >>On December 30, 2005 at 13:02:43, Ingo Althofer wrote: >> in case of linear evaluation functions with lots >>>of terms there is always a small subset of the terms >>>such that this set with the right parameters is >>>almost as good as the full evaluation function. >> >>The whole multivariate approach to the social sciences, of which multiple linear >>regression is an example, is based on the same assumption. > ^^^^^^^^^^^^^^^^ > >My result mentioned above is not an assumption, but >a proof (with arguments from linear algebra and >probability theory). > >But, of course, it is not an accidence that the >principal component analysis works so well in >many (also social) sciences. But nobody ever could _prove_ such some for social sciences. :) > > >>Parsimony rules. > > >Or, formulated more positively: >"Occams razor is the best we have!" > >Ingo.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.