pcp draws a parallel coordinate plot.It is a modified
version of parcoord {MASS}.
Variables
may be reordered and panels colored in the display.
catpcp draws a parallel coordinate plot variant for categorical data.
pcp(data, order = NULL, panel.colors = NULL, col = 1, lty = 1,
horizontal = TRUE, mar = NULL, scale = TRUE, axis.width = 0,
axis.grid.col="grey70",connect=TRUE, ...)catpcp(data, order = NULL, pcpbars, barvars = 1:ncol(data),
pcpbars.border = "black", pcpbars.col = NULL, pcpbars.labels = FALSE,
pcpbars.axis.at = NULL, pcpbars.axis.labels = NULL,
axis.width = 0.2,connect=TRUE, ...)
A data frame or matrix.
an index vector specifying variable order. If NULL, all variables are used.
either a vector or a matrix of panel colors. If a vector is supplied, the ith color is used for the ith panel. If a matrix, dimensions should match those of the variables. Diagonal entries are ignored.
a vector of colours, recycled as necessary for each observation.
a vector of line types, recycled as necessary for each observation.
If TRUE, orientation is horizontal.
margin parameters, passed to par.
If TRUE, the variables are scaled to the unit interval.
Width of each of the parallel axes.
Color of variable axes. Use NULL for no axes.
If FALSE, line segments are not connected. Points are drawn if axis.width=0.
A list, with one component per barvar. Component i is a matrix with the bottom and top of the bars for that variable.
Categorical variables where overlayed bars show the level frequency.
Border colour of the bars.
Colour of the bars.
Labels for the bars.
Axis label positions for the bars.
Axis label text for the bars.
other parameters, passed to pcp by catpcp
require(PairViz)
y <- as.data.frame(as.table(HairEyeColor))
colvar <- 3 # any of 1:3 will do
y <- y[order(y[,colvar]),] # ensures that cases are ordered by colour within each factor level
ylong <- apply(y[,-4],2, function(x) rep(x,times=y[,4]))
cols <- desaturate_color(rainbow(4,alpha=0.3),.5)
cols <- cols[as.numeric(as.factor(ylong[,colvar]))]
ds <- factor_spreadout(ylong)
dev.new(width=5,height=2.5)
par(mar=c(2,1,2,1))
par( cex.axis=.8,cex=.8)
catpcp(ds$data,col=cols,lwd=2,pcpbars=ds$bars,pcpbars.labels=TRUE,main="Hair Eye data")
Run the code above in your browser using DataLab