[R] inter-specific competition - community matrices and two species models using Lotka-Volterra

Ben Bolker bbolker at gmail.com
Sun Feb 20 23:41:21 CET 2011


Bob O'Hara <rni.boh <at> gmail.com> writes:

> 
> On 20 February 2011 16:19, Chris Buddenhagen wrote:
> 
> > Does anyone know of example r-code/packages for carrying out analysis?
> > Preferably this would have examples from real experimental data of two or
> > more competing species...
> >
> > In discrete time, you can use a Gompertz model:
> <http://rspb.royalsocietypublishing.org/content/276/1669/2923>
> which reduces to a VAR(1) model, which is a multivariate time series model,
> check here:
> <http://cran.r-project.org/web/views/TimeSeries.html>
> 

  If you have observation error (which you might be able to ignore/
say is negligible in some *experimental* contexts, although hardly
ever in observational contexts) you would want to take this one
step further and think about using the MARSS (multivariate 
autoregressive state-space) package (also check out dlm, sspir packages).
  You should definitely take a look at Ives et al 2003 Ecological
Monographs: unfortunately for you, they used Matlab rather than
R, but you might be able to make the translation to the MARSS package
fairly painlessly since the underlying methods are very very similar.



More information about the R-help mailing list