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 #55712 is a reply to message #55697]

Tue, 04 September 2007 09:23

David Fanning
Messages: 11724
Registered: August 2001

Senior Member

Jeff N. writes:

> 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).

Oh, my goodness. I must have been asleep when I wrote that.
One of the problems with working at 3AM. :-(

Fixed now. Thanks.

Cheers,

David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.dfanning.com/

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: Mon Dec 01 12:47:21 PST 2025

Total time taken to generate the page: 0.25072 seconds