splineGradient: Does post-processing of canonical analysis object from vegan...

In rbmillar/SplineCAP: What the Package Does (One Line, Title Case)

Description

Does post-processing of canonical analysis object from vegan to extract spline gradients

Usage

splineGradient(object, Bdf, Gnames = NULL, predB = NULL, plot = T, ...)

Arguments

object	The fitted rda/cca or capscale object
Bdf	Integer vector giving the degrees of freedom (number of columns) of the bases used.
Gnames	Gradient names corresponding to the bases.
predB	List containing the new set of bases for prediction.
plot	Draw plot of the gradient. predB must be supplied.
...	Arguments to be passed on to the plot command.

Value

Gradt Fitted gradient values

Env.scores Correlation scores for spline and linear covariates, for adding to a biplot

pred.Gradt List containing the fitted gradient values for predB

Examples

data(PitTraps)
Locn=PitTraps$Locn
distance=PitTraps$distance
Counts=PitTraps[,-c(1,2)]
B=bSpline(distance,degree=2,df=5)   #Spline basis for distance
Bpred=predict(B,newx=min(distance):max(distance)) #For prediction
CAP.fit=capscale((Counts>0)~B+Locn,dist="bray") #Fit RDA on PCOs
Gradients=splineGradient(CAP.fit,Bdf=5,predB=list(Bpred),type="l")

#Draw triplot
env=Gradients$Env.scores
ordiplot(CAP.fit,display = c("sp", "wa", "cn")[1:2],scaling=3,
            cex=0.6,xlim=c(-1.5,1.5),ylim=c(-1.5,1.5))
arrows(0,0,env[1,],env[2,],length=0.05,col="blue",lwd=1.5)
text(env[1,],env[2,],colnames(env),cex=0.8,adj=c(0,0.5))

rbmillar/SplineCAP documentation built on Feb. 20, 2021, 6:40 a.m.