PlotMatrix(x, y = NULL, data = NULL, panel = l.panel, nrows = 0, ncols = nrows, save = TRUE, robrange. = FALSE, range. = NULL, pch = NULL, col = 1, reference = 0, ltyref = 3, log = "", xaxs = "r", yaxs = "r", xaxmar = NULL, yaxmar = NULL, vnames = NULL, main = "", cex.points = NA, cex.lab = 0.7, cex.text = 1.3, cex.title = 1, bty = "o", oma = NULL, ...)
x
x
or y
is a formulasave==TRUE
,
the first row and the last column are suppressed.colnames
identifying the variables."x"
asks for
log scale on horizontal axis, "y"
, on vertical axis,
"xy"
, on both axes.par
cex
par
par
panel
functionx
or y
is a data.frame, it is converted to a
numerical matrix. The panel
function can be user written. It needs $>=6$ arguments,
which are given:
y
variables,
x
variables,
pch
, and
col
Since large scatterplot matrices lead to tiny panels, PlotMatrix
splits the matrix into blocks of at most nrows
rows and
ncols
columns. If these numbers are missing, they default to
nrows=5
and ncols=6
for landscape pages, and to
nrows=8
and ncols=5
for portrait pages.
pairs
PlotMatrix(iris[,1:4], main="Iris", pch=as.numeric(iris[,"Species"]))
Run the code above in your browser using DataLab