mstree {ade4}R Documentation

Minimal Spanning Tree

Description

Minimal Spanning Tree

Usage

mstree(xdist, ngmax = 1)

Arguments

xdist

an object of class dist containing an observed dissimilarity

ngmax

a component number (default=1). Select 1 for getting classical MST. To add n supplementary edges k times: select k+1.

Value

returns an object of class neig

Author(s)

Daniel Chessel

Examples

data(mafragh)
maf.coa <- dudi.coa(mafragh$flo, scan = FALSE)
maf.mst <- ade4::mstree(dist.dudi(maf.coa), 1)

if(adegraphicsLoaded()) {
  g0 <- s.label(maf.coa$li, plab.cex = 0, ppoints.cex = 2, nb = neig2nb(maf.mst))
} else {
  s.label(maf.coa$li, clab = 0, cpoi = 2, neig = maf.mst, cnei = 1)
}

xy <- data.frame(x = runif(20), y = runif(20))

if(adegraphicsLoaded()) {
  g1 <- s.label(xy, xlim = c(0, 1), ylim = c(0, 1), 
    nb = neig2nb(ade4::mstree(dist.quant(xy, 1), 1)), plot = FALSE)
  g2 <- s.label(xy, xlim = c(0, 1), ylim = c(0, 1), 
    nb = neig2nb(ade4::mstree(dist.quant(xy, 1), 2)), plot = FALSE)
  g3 <- s.label(xy, xlim = c(0, 1), ylim = c(0, 1), 
    nb = neig2nb(ade4::mstree(dist.quant(xy, 1), 3)), plot = FALSE)
  g4 <- s.label(xy, xlim = c(0, 1), ylim = c(0, 1), 
    nb = neig2nb(ade4::mstree(dist.quant(xy, 1), 4)), plot = FALSE)
  G <- ADEgS(list(g1, g2, g3, g4), layout = c(2, 2))

} else {
  par(mfrow = c(2, 2))
  for(k in 1:4) {
    neig <- mstree(dist.quant(xy, 1), k)
    s.label(xy, xlim = c(0, 1), ylim = c(0, 1), addax = FALSE, neig = neig)
  }
}

[Package ade4 version 1.7-4 Index]