Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.soft-sys.math.mathematica > #16107 > unrolled thread

Re: Principal Components

Started byBob Hanlon <hanlonr357@gmail.com>
First post2013-10-24 03:43 +0000
Last post2013-10-24 03:43 +0000
Articles 1 — 1 participant

Back to article view | Back to comp.soft-sys.math.mathematica

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: Principal Components Bob Hanlon <hanlonr357@gmail.com> - 2013-10-24 03:43 +0000

#16107 — Re: Principal Components

FromBob Hanlon <hanlonr357@gmail.com>
Date2013-10-24 03:43 +0000
SubjectRe: Principal Components
Message-ID<l4a50v$ags$1@smc.vnet.net>
Covariance is the default method


$Version


"9.0 for Mac OS X x86 (64-bit) (January 24, 2013)"


Options[PrincipalComponents]


{Method -> "Covariance"}


mat = {{1., 2}, {2, 3}, {4, 10}};


pc1 = PrincipalComponents[mat]


{{3.27053, 0.285293}, {1.99969, -0.335165}, {-5.27023, 0.0498715}}


pc1 == PrincipalComponents[mat,
  Method -> "Covariance"] ==
 PrincipalComponents[pc1] ==
 PrincipalComponents[pc1,
  Method -> "Covariance"]


True


pc2 = PrincipalComponents[mat,
  Method -> "Correlation"]


{{1.10388, 0.130549}, {0.478746, -0.170139}, {-1.58262, 0.0395904}}


pc2 == PrincipalComponents[pc2] ==
 PrincipalComponents[pc2,
  Method -> "Covariance"]


True



Bob Hanlon




On Tue, Oct 22, 2013 at 12:49 AM, pw <p.willis@telus.net> wrote:

>
> Hello,
>
> The function 'PrincipalComponents' returns the the first principal
> components vector of a group of input vectors.
>
> Question 1.) Is it possible to define Method->"covariance" or is
> correlation the only PC technique available?
>
> Question 2.) How can I generate an inverse PC transform from my
> PrincipalComponents output?
>
>
> Thanks,
>
>
> Peter
>
>

[toc] | [standalone]


Back to top | Article view | comp.soft-sys.math.mathematica


csiph-web