First of all, thanks for the answers. I dont know if my first question was right, anyway I would
like know how I could get
the percentage of total variation
explained by each CV. For example, in Astúa 2009 [ Evolution of scapula size
and shape in Didelphid marsupials (Didelphimorphia: Didelphidae). Evolution
63-9:2438-2456], in Figures 4,5 & 6 are represented some CVA obtained from
CVAGen6 and in their axis (and on the manuscript) the percentage of variance of
CVs are given. I was talking about of
that
percentage and how to obtain it because with CVAGen6 I couldn’t manage. I did the regression of shape variables on CV scores with
tpsRegr and I visualized the shape deformation implied by changes along each
axis. Is the “percent unexplained” value that the tpsRegr report give this percenatge? I hope in this time have been more clear. Thanks again. Beatriz Gamarra
----- Forwarded message from
beatriz gamarra
Date: Thu, 7 Jun 2012
12:52:08 -0400
From: beatriz gamarra
Reply-To: beatriz gamarra
Subject: Re: CV shape
variation
To: morphmet@morphometrics.org
-------- Original Message --------
Subject: RE: CV shape variation
Date: Mon, 4 Jun 2012 15:05:53 -0400
From: F. James Rohlf <ro...@life.bio.sunysb.edu>
Reply-To: ro...@life.bio.sunysb.edu
Organization: Stony Brook University
To: morphmet@morphometrics.org
I am not sure of the question being asked. Perhaps what was meant was
how to visualize the shape deformation implied by changes along each
axis? I think the CVAGen6 software has an option for that. Basically,
one regresses shape on the CVA scores. This can be done in the tpsRegr
software.
----------------------
F. James Rohlf, John S. Toll Professor, Stony Brook University
The much revised 4^th editions of Biometry and Statistical Tables are
now available:
http://www.whfreeman.com/Catalog/product/biometry-fourthedition-sokal
http://www.whfreeman.com/Catalog/product/statisticaltables-fourthedition-rohlf
PPlease consider the environment before printing this email
*From:*morphmet_moderator@morphometrics.org
[mailto:morphmet_moderator@morphometrics.org]
*Sent:* Monday, June 04, 2012 2:03 AM
*To:* morphmet@morphometrics.org
*Subject:* Re: CV shape variation
----- Forwarded message from Philipp Mitteröcker -----
Date: Fri, 1 Jun 2012 05:52:44 -0400
From: Philipp Mitteröcker
Reply-To: Philipp Mitteröcker
Subject: Re: CV shape variation
For a single canonical variate, you can compute the variance along this
CV and divide it by the total variance (computed, e.g., as the sum of
the variances of all variables). But in contrast to PCA, you cannot add
up the variances for two or more CVs, because the CV axes are not
orthogonal. You could compute the variance within the plane spanned by
two CVs when orthogonalizing the two CV axes.
However, explained variance among the individuals is not often reported
for CVA, as it is not the quantity that is maximized.
Best,
Philipp
Am 01.06.2012 um 06:27 schrieb morphmet_moderator@morphometrics.org
----- Forwarded message from beatriz gamarra-----
Date: Wed, 30 May 2012 13:29:58 -0400
From: beatriz gamarra
Reply-To: beatriz gamarra
Subject: CV shape variation
To:morphmet-request@morphometrics.org
Hi,
I amanalyzinga dataset with 12 landmarks and 12 groups. I have run CVA
by using CVAGen6 from IMP and I want to obtain the shape variation
associated with first CVs. How could I achieve?
Thanks in advance.
Beatriz Gamarra
----- End forwarded message -----
___________________________________
Dr. Philipp Mitteroecker
Department of Theoretical Biology
University of Vienna
Althanstrasse 14
A-1090 Vienna, Austria
Tel: +43 1 4277
56705
Fax: +43 1 4277
9544
email: philipp.mitteroecker@univie.ac.at
----- End forwarded message
-----
2012/6/7 morphmet <morphmet_modera...@morphometrics.org>
- Fwd: CV shape variation morphmet_moderator
- Re: CV shape variation morphmet_moderator
- RE: CV shape variation morphmet
- Re: CV shape variation morphmet_moderator