Scatterplot Matrices

A matrix of scatterplots is produced.

pairs(x, ...)
"pairs"(formula, data = NULL, ..., subset, na.action = stats::na.pass)
"pairs"(x, labels, panel = points, ..., lower.panel = panel, upper.panel = panel, diag.panel = NULL, text.panel = textPanel, label.pos = 0.5 + has.diag/3, line.main = 3, cex.labels = NULL, font.labels = 1, row1attop = TRUE, gap = 1, log = "")
the coordinates of points given as numeric columns of a matrix or data frame. Logical and factor columns are converted to numeric in the same way that data.matrix does.
a formula, such as ~ x + y + z. Each term will give a separate variable in the pairs plot, so terms should be numeric vectors. (A response will be interpreted as another variable, but not treated specially, so it is confusing to use one.)
a data.frame (or list) from which the variables in formula should be taken.
an optional vector specifying a subset of observations to be used for plotting.
a function which indicates what should happen when the data contain NAs. The default is to pass missing values on to the panel functions, but na.action = na.omit will cause cases with missing values in any of the variables to be omitted entirely.
the names of the variables.
function(x, y, ...) which is used to plot the contents of each panel of the display.
arguments to be passed to or from methods.

Also, graphical parameters can be given as can arguments to plot such as main. par("oma") will be set appropriately unless specified.

lower.panel, upper.panel
separate panel functions (or NULL) to be used below and above the diagonal respectively.
optional function(x, ...) to be applied on the diagonals.
optional function(x, y, labels, cex, font, ...) to be applied on the diagonals.
y position of labels in the text panel.
if main is specified, line.main gives the line argument to mtext() which draws the title. You may want to specify oma when changing line.main.
cex.labels, font.labels
graphics parameters for the text panel.
logical. Should the layout be matrix-like with row 1 at the top, or graph-like with row 1 at the bottom?
distance between subplots, in margin lines.
a character string indicating if logarithmic axes are to be used: see plot.default. log = "xy" specifies logarithmic axes for all variables.

The $ij$th scatterplot contains x[,i] plotted against x[,j]. The scatterplot can be customised by setting panel functions to appear as something completely different. The off-diagonal panel functions are passed the appropriate columns of x as x and y: the diagonal panel function (if any) is passed a single column, and the text.panel function is passed a single (x, y) location and the column name. Setting some of these panel functions to NULL is equivalent to not drawing anything there.

The graphical parameters pch and col can be used to specify a vector of plotting symbols and colors to be used in the plots.

The graphical parameter oma will be set by pairs.default unless supplied as an argument.

A panel function should not attempt to start a new plot, but just plot within a given coordinate system: thus plot and boxplot are not panel functions.

By default, missing values are passed to the panel functions and will often be ignored within a panel. However, for the formula method and na.action = na.omit, all cases which contain a missing values for any of the variables are omitted completely (including when the scales are selected).


Becker, R. A., Chambers, J. M. and Wilks, A. R. (1988) The New S Language. Wadsworth & Brooks/Cole.

  • pairs
  • pairs.default
  • pairs.formula
library(graphics) pairs(iris[1:4], main = "Anderson's Iris Data -- 3 species", pch = 21, bg = c("red", "green3", "blue")[unclass(iris$Species)]) ## formula method pairs(~ Fertility + Education + Catholic, data = swiss, subset = Education < 20, main = "Swiss data, Education < 20") pairs(USJudgeRatings) ## show only lower triangle (and suppress labeling for whatever reason): pairs(USJudgeRatings, text.panel = NULL, upper.panel = NULL) ## put histograms on the diagonal panel.hist <- function(x, ...) { usr <- par("usr"); on.exit(par(usr)) par(usr = c(usr[1:2], 0, 1.5) ) h <- hist(x, plot = FALSE) breaks <- h$breaks; nB <- length(breaks) y <- h$counts; y <- y/max(y) rect(breaks[-nB], 0, breaks[-1], y, col = "cyan", ...) } pairs(USJudgeRatings[1:5], panel = panel.smooth, cex = 1.5, pch = 24, bg = "light blue", diag.panel = panel.hist, cex.labels = 2, font.labels = 2) ## put (absolute) correlations on the upper panels, ## with size proportional to the correlations. panel.cor <- function(x, y, digits = 2, prefix = "", cex.cor, ...) { usr <- par("usr"); on.exit(par(usr)) par(usr = c(0, 1, 0, 1)) r <- abs(cor(x, y)) txt <- format(c(r, 0.123456789), digits = digits)[1] txt <- paste0(prefix, txt) if(missing(cex.cor)) cex.cor <- 0.8/strwidth(txt) text(0.5, 0.5, txt, cex = cex.cor * r) } pairs(USJudgeRatings, lower.panel = panel.smooth, upper.panel = panel.cor) pairs(iris[-5], log = "xy") # plot all variables on log scale pairs(iris, log = 1:4, # log the first four main = "Lengths and Widths in [log]", line.main=1.5, oma=c(2,2,3,2))
Documentation reproduced from package graphics, version 3.0.3, License: Part of R 3.0.3

Community examples

Looks like there are no examples yet.