0
$\begingroup$

I would like to write certain model in R.

I would like to o use a vector object as a dependent variable in my general linear model. The model is supposed to explain how some environmental factors, and their interactions, explain the shape of a species community composition.

The vector object is supposed to look something like [0,2,0,3,1] and designate species community composition. Each dimension of the vector is intended to be a different species, and their values are intended to show the abundance of species.

Does anybody know in which package is it possible to do it/ is it possible at all?

If anybody has a better idea how to model such a relation, I am also open to ideas ;)

Thanks in advance,

Best wishes,

Jacek

$\endgroup$
0
$\begingroup$

I don't think it's possible to use a vector as dependent variable for a linear model.

First option, if you really want to link abiotic variables to community with a model, is to calculate community metrics (e.g. Shannon Index, Species Richness, biological traits, etc.) and use each metric as dependent variable in models with your environmental factors as explanatory variables. An alternative can be using variables linked to a unique species (e.g. abundance) as dependent variable but I am pretty sure you can't use a vector of abundance in a linear model. And you won't link community composition to environmental factors, you will only link the abundance of one species (or one community metric) to these factors.

Second option, what I would recommand here, is to use multivariate methods. Canonical Correspondence Analysis might be relevent in your case but you should read some papers (e.g. this excellent article) and check this great website about multivariate methods in ecology before you pick a procedure.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.