swapClass-package: SwapClass : a simple null model adapted to abundance classes...

In swapClass: A Null Model Adapted to Abundance Class Data in Ecology

Description

A null model randomizing semi-quantitative multi-classes (or ordinal) data by swapping sub-matrices while both the row and the column marginal sums are held constant.

Details

Package:	swapClass
Type:	Package
Version:	1.0.1
Date:	2017-06-22
License:	GPL-3

Author(s)

Benjamin Borgy ([e-mail]), Remi Perronne, Xavier Reboud

References

Borgy B. - Dynamic and assembly of weed communities: Approach by statistical modeling (2011). Ph.D. Thesis. INRA Agroicologie & Universiti de Bourgogne. DOI: 10.13140/RG.2.1.1738.1601

Examples

MAT=matrix(sample(0:3,50*20,replace=TRUE,prob=c(.7,.1,.1,.1)),ncol=20)

#Calculation of the nbPerm index
nbPerm(MAT)

#Generation of null communities
NULL_MATS=nullModel(MAT)

#First null community
NULL_MATS$sim[[1]]

#Number of times that each cell has been swapped for the first null community
NULL_MATS$perms[[1]]

#nbPerm index over the observed community and the 100 null communities
plot(c(nbPerm(MAT),unlist(lapply(NULL_MATS$sim,nbPerm))),type='l',ylab="nbPerm index")

#the number of each classes per row are equal
#between observed community and the first null community
f_table = function(x) table(factor(x,levels=0:3))
all(apply(MAT,1,f_table)==apply(NULL_MATS$sim[[1]],1,f_table))

#the number of each classes per column are equal
#between observed community and the first null community
all(apply(MAT,2,f_table)==apply(NULL_MATS$sim[[1]],2,f_table))

swapClass documentation built on May 1, 2019, 7:24 p.m.