kplot.sepan {ade4}R Documentation

Multiple Graphs for Separated Analyses in a K-tables

Description

performs high level plots for Separed Analyses in a K-tables, using an object of class sepan.

Usage

## S3 method for class 'sepan':
kplot(object, xax = 1, yax = 2, which.tab = 1:length(object$blo), 
    mfrow = NULL, permute.row.col = FALSE, clab.row = 1, 
    clab.col = 1.25, traject.row = FALSE, csub = 2, 
    possub = "bottomright", show.eigen.value = TRUE,...)

## S3 method for class 'sepan.coa':
kplot(object, xax = 1, yax = 2, which.tab = 1:length(object$blo), 
    mfrow = NULL, permute.row.col = FALSE, clab.row = 1, 
    clab.col = 1.25, csub = 2, possub = "bottomright", 
    show.eigen.value = TRUE, poseig = c("bottom", "top"), ...)

Arguments

object an object of class sepan
xax, yax the numbers of the x-axis and the y-axis
which.tab a numeric vector containing the numbers of the tables to analyse
mfrow parameter for the array of figures to be drawn, otherwise use n2mfrow
permute.row.col if TRUE the rows are represented by arrows and the columns by points, if FALSE it is the opposite
clab.row a character size for the row labels
clab.col a character size for the column labels
traject.row a logical value indicating whether the trajectories between rows should be drawn in a natural order
csub a character size for the sub-titles, used with par("cex")*csub
possub a string of characters indicating the sub-title position ("topleft", "topright", "bottomleft", "bottomright")
show.eigen.value a logical value indicating whether the eigenvalues bar plot should be drawn
poseig if "top" the eigenvalues bar plot is upside, if "bottom", it is downside
... further arguments passed to or from other methods

Details

kplot.sepan superimposes the points for the rows and the arrows for the columns using an adapted rescaling such as the scatter.dudi.
kplot.sepan.coa superimposes the row coordinates and the column coordinates with the same scale.

Author(s)

Daniel Chessel

Examples

data(escopage)
w <- data.frame(scale(escopage$tab))
w <- ktab.data.frame(w, escopage$blo, tabnames = escopage$tab.names)
sep1 <- sepan(w)
kplot(sep1, show = FALSE)

data(friday87)
w <- data.frame(scale(friday87$fau, scal = FALSE))
w <- ktab.data.frame(w, friday87$fau.blo, tabnames = friday87$tab.names)
kplot(sepan(w), clab.r = 1.25, clab.c = 0, csub = 3)

data(microsatt)
w <- dudi.coa(data.frame(t(microsatt$tab)), scann = FALSE)
loci.fac <- factor(rep(microsatt$loci.names, microsatt$loci.eff))
wit <- within(w, loci.fac, scann = FALSE)
microsatt.ktab <- ktab.within(wit)
kplot.sepan.coa(sepan(microsatt.ktab), show = FALSE, clab.c = 0, 
    mfrow = c(3,3), clab.r = 1.5)

Worked out examples


> library(ade4)
> ### Name: kplot.sepan
> ### Title: Multiple Graphs for Separated Analyses in a K-tables
> ### Aliases: kplot.sepan kplot.sepan.coa
> ### Keywords: multivariate hplot
> 
> ### ** Examples
> 
> data(escopage)
> w <- data.frame(scale(escopage$tab))
> w <- ktab.data.frame(w, escopage$blo, tabnames = escopage$tab.names)
> sep1 <- sepan(w)
> kplot(sep1, show = FALSE)
> 
> data(friday87)
> w <- data.frame(scale(friday87$fau, scal = FALSE))
> w <- ktab.data.frame(w, friday87$fau.blo, tabnames = friday87$tab.names)
> kplot(sepan(w), clab.r = 1.25, clab.c = 0, csub = 3)
> 
> data(microsatt)
> w <- dudi.coa(data.frame(t(microsatt$tab)), scann = FALSE)
> loci.fac <- factor(rep(microsatt$loci.names, microsatt$loci.eff))
> wit <- within(w, loci.fac, scann = FALSE)
> microsatt.ktab <- ktab.within(wit)
> kplot.sepan.coa(sepan(microsatt.ktab), show = FALSE, clab.c = 0, 
+     mfrow = c(3,3), clab.r = 1.5)
> 
> 
> 
> 

[Package ade4 Index]