Re: [R] vegan CCA I am Completely new to ordination analyses

From: Jari Oksanen <jari.oksanen_at_oulu.fi>
Date: Mon, 14 Mar 2011 06:25:46 +0000

kitty <kitty.a1000 <at> gmail.com> writes:

>
> Dear list,
>
> I am trying to predict species volume from bioclimatic data, I have various
> sites and I have a data frame with species volume and
> the corresponding bioclimatic data for each site.
>
> I read on a discussion forum that you can use ordination to predict species
> abundance (in my case volume) from 'new' climate data for sites where you do
> not know the abundance.
>
> Unfortunately I cannot work out how to do this in the r package vegan, in
> fact I cannot even get a simple CCA to work; I have attached a sample of my
> data (volume and bioclimate data for around 100 of my sites) and the code I
> used is below.
>
> Does the output look right? why am I getting a 0 output? Is
> there something wrong with my data, data format, r code etc?
>
Ordination, including cca(), is multivariate analysis. If you only have one dependent variate, you should use univariate methods (regression etc.). Some multivariate methods fall back to univariate methods if you only have one dependent variable, but cca() does not do so in a meaningful way. In fact cca() does do this, but not meaningfully: your residual variation is zero and rank is zero. Switch to regression.

Cheers, Jari Oksanen



R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Mon 14 Mar 2011 - 06:30:07 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Mon 14 Mar 2011 - 07:30:20 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive