comp.lang.idl-pvwave archive: archive » Re: Principal Componets Analysis

Home » Public Forums » archive » Re: Principal Componets Analysis

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: Principal Componets Analysis [message #55714 is a reply to message #55712]

Tue, 04 September 2007 09:02

Jeff N.
Messages: 120
Registered: April 2005

Senior Member

David,

I'm reading through the tutorial, and spotted something you might want
to fix in your discussion of CORRELATE. It's not true that negative
values returned from that function are uncorrelated. You actually
describe the reason why this is true in your tutorial: "a change in
one vector will predict an opposite change in the other." if one
vector predicts a change in the other, it's correlated. The negative
sign just means that the change is an opposite change, like you
mentioned. How well two variables are correlated depends on the
magnitude of the number: 0 is not correlated, 1 is perfectly
correlated (so -1 is perfectly negatively correlated).

Jeff

Report message to a moderator

[Message index]

		Re: Principal Componets Analysis By: wlandsman on Wed, 05 September 2007 07:45
		Re: Principal Componets Analysis By: David Fanning on Tue, 04 September 2007 09:23
		Re: Principal Componets Analysis By: Jeff N. on Tue, 04 September 2007 09:02
		Re: Principal Componets Analysis By: David Fanning on Mon, 03 September 2007 18:06

Previous Topic:	TIFF Data formats
Next Topic:	Re: TIFF Data formats

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 19:25:42 PDT 2025

Total time taken to generate the page: 0.00403 seconds