fanovaGraph (version 1.5)

plot.graphlist: Plot Graph via Package igraph

Description

Plot FANOVA graphs using functions from package igraph.

Usage

# S3 method for graphlist
plot(x, names = NULL, i2 = NULL, layout = NULL, plot.i1=TRUE, max.thickness=15, 
circle.diameter=40, ...)

Arguments

x

an object of class graphlist as obtained from estimateGraph

names

optional character string, names of vertices, defaults to 1:d

i2

optional vector of second order interaction indices (thickness of inner edges)

plot.i1

optional boolean, if TRUE main effects are drawn in the graph by vertices thicknesses, should be set to FALSE when only total interaction indices are of interest

layout

optional layout for the graph as in igraph, default is layout.fruchterman.reingold

max.thickness

optional value for the maximal line thickness, defaults to 20

circle.diameter

optional value for the circle diameter, defaults to 40

...

additional arguments, passed to plot

References

Muehlenstaedt, T.; Roustant, O.; Carraro, L.; Kuhnt, S. (2011) Data-driven Kriging models based on FANOVA-decomposition, Statistics and Computing.

Csardi, G.; Nepusz, T. (2006) The igraph software package for complex network research, InterJournal Complex Systems, Complex Systems, 1695.

See Also

plotGraphChange

Examples

# NOT RUN {
op <- par(no.readonly=TRUE)
g1 <- estimateGraph(f.mat=ishigami.fun, d=3, q.arg=list(min=-pi,max=pi), n.tot=10000)
plot(g1)
plot(g1, names=c("A","B","C"))
plot(g1, names=c("A","B","C"), plot.i1 = FALSE)

# include pure second order indices
g2 <- estimateGraph(f.mat=function(x) x[,1]*x[,2]*x[,3]+x[,2]*x[,3], d=3, 
  q.arg=list(min=-1,max=1), n.tot=10000)

plot(g2)
plot(g2, i2 = c(0.001, 0.001, 0.05))

# equal layouts and different edge thicknesses and circle diameters
g3 <- estimateGraph(f.mat=function(x) x[,1]*x[,2]*x[,3]*x[,4]*x[,5], d=5, 
  q.arg=list(min=-1,max=1), n.tot=10000)

g4 <- estimateGraph(f.mat=function(x) x[,1]*x[,2]*x[,3]+x[,4]*x[,5], d=5, 
  q.arg=list(min=-1,max=1), n.tot=10000)

graphClassIgraph <- graph.full(n = 5, directed = FALSE)
layout <- layout.circle(graphClassIgraph)

plot(g3, max.thickness= 10, circle.diameter= 30, layout=layout)
plot(g4, max.thickness= 30, circle.diameter= 50, layout=layout)
# }