-------- Original Message --------
Subject: RE: R covariance matrix and eigenvalues...
Date: Thu, 23 Sep 2010 19:40:43 -0400
From: Sam Brown <[email protected]>
To: <[email protected]>
Gidday Milos
Not sure whether you're wanting intellegent comment on Wagner's methods
specifically, or a more general guide to matrix algebra in R. As I don't
know Wagners methods, I am going to answer the latter.
R provide a number of functions for dealing with matrices, including the
calculation of covariance matrices and eigenvalues as shown below.
Here's a couple of introductions to matrix algebra in R:
http://personality-project.org/r/sem.appendix.1.pdf
http://www.statpower.net/Content/MLRM/Lecture%20Slides/RMatrixIntro.pdf
Also, the help file for any function can be viewed by typing a ? before
the function name e.g. ?eigen
Hope this helps!
Cheers
Samuel Brown
Casual Research Assistant
Bio-Protection Research Centre
PO Box 84
Lincoln University
Lincoln 7647
Canterbury
New Zealand
> x<-matrix(1:12, ncol=3)
> x
[,1] [,2] [,3]
[1,] 1 5 9
[2,] 2 6 10
[3,] 3 7 11
[4,] 4 8 12
> cov(x)
[,1] [,2] [,3]
[1,] 1.666667 1.666667 1.666667
[2,] 1.666667 1.666667 1.666667
[3,] 1.666667 1.666667 1.666667
> eigen(cov(x))
$values
[1] 5.000000e+00 2.664535e-15 3.409816e-16
$vectors
[,1] [,2] [,3]
[1,] 0.5773503 0.8164966 0.0000000
[2,] 0.5773503 -0.4082483 -0.7071068
[3,] 0.5773503 -0.4082483 0.7071068
A link to the Wagner paper is available here, though I don't have access
to it unfortunately.
http://onlinelibrary.wiley.com/doi/10.1111/j.1439-0469.1990.tb00364.x/abstract
> Date: Wed, 22 Sep 2010 07:55:21 -0400
> From: [email protected]
> To: [email protected]
> Subject: R covariance matrix and eigenvalues...
>
>
>
> -------- Original Message --------
> Subject: R covariance matrix and eigenvalues...
> Date: Tue, 21 Sep 2010 11:38:39 +0100
> From: Milos Blagojevic <[email protected]>
> To: <[email protected]>
>
>
>
> Hello to all morphometricians,
>
> I am wondering about some methodological issues concerning
> implementation of matrix correlations for landmark data in R. All papers
> that I have read are pointing to the work of Wagner
> "A comparative study of morphological integration in Apis mellifera" and
> the scaled eigenvalues of covariance matrices as
> a measure of morphological integration. A good set of matrix correlation
> tools are contained in Marquez`s Mace software, but I am not completely
> sure how all this should be done. I wanted
> to rework through it using R, but I lack experience and maths knowledge.
> Also I can`t find this paper from Wagner anywhere. Have anyone had
> experience with this type of analyses in R and
> can share some advice with me?Maybe a chat session via skype or
google talk?
>
>
> Every advice is welcomed and most sought-for,
> best regards
>
> Milos Blagojevic,PhD student
> Faculty of science
> Depatment for biology and ecology
> Radoja Domanovica 12
> 34000 Kragujevac
> Serbia
> e-mail:[email protected] - morphmet
> [email protected]
>
>
