CoMLsimulation: Simulated site-species data

In slarge/gradientForest: Random Forest functions for the Census of Marine Life synthesis project.

Description

Simulated datasets for demonstrating gradient forest method.

Usage

data(CoMLsimulation)

Details

Xsimulation

A data frame with 100 observations on 10 variables.

A, B

influential variables, each generated uniformly on [0-1]

C-J

uninfluential variables, each generated uniformly on [0-1]

Ysimulation

A matrix of simulated species counts on the 100 sites in Xsimulation. The species a1-a3 respond to variable A, b1-b4 respond to variable B and species ab1-ab5 respond jointly to both A and B.

The species data are generated as Poissons with intensity shaped like normal curves along A and B gradients.

References

Ellis, N., Smith, S.J., and Pitcher, C.R. (2012) Gradient Forests: calculating importance gradients on physical predictors. Ecology, 93, 156–168.

Examples

data(CoMLsimulation)
names_a <- paste("a",1:3,sep="")
names_b <- paste("b",1:4,sep="")
par(mfrow=c(2,1),mar=c(3,4,3,1))
matplot(Xsimulation$A,Ysimulation[,names_a],main=substitute(a[1-3]),xlab="A",
ylab="Abundance",axes=FALSE,col=c("black","orange","blue","green"))
mtext("A",side=1,line=1)
axis(2)
box()
matplot(Xsimulation$B,Ysimulation[,names_b],main=substitute(b[1-4]),xlab="B",
ylab="Abundance",axes=FALSE,col=c("black","orange","blue","green"))
mtext("B",side=1,line=1)
axis(2)
box()

slarge/gradientForest documentation built on May 3, 2019, 4:05 p.m.