This function is generic (see Methods); method functions can be written to handle specific classes of data. Classes which already have methods for this function include: princomp, factanal.

biplot(x, ...)
biplot.default(obs, bivars, var.axes=T, col=<<see below>>,
     cex=<<see below>>, olab=NULL, vlab=NULL, expand=1,
     xlim=NULL, ylim=NULL)

x:

object having a class for which a method exists.

obs:

two-column matrix of observations.

bivars:

two-column matrix of (scaled) variables.

var.axes:

a logical vector giving the variables for which arrows should be drawn. This may have names, which are (partially) matched against the names of the variables, or without names it is replicated to have length equal to the number of variables.

col:

vector of length 1 or 2 stating which colors to use for the observations and for the variables. The default is par("col") for the observations and par("col")+1 for the variables.

cex:

vector of length 1 or 2 giving the cex parameter for the observations and for the variables. The default is to use the current cex for both.

olab:

vector of character strings which label the observations. This will be replicated to be the proper length. The default is to use the dimnames from obs if present, and the observation number otherwise.

vlab:

vector of character strings which label the variables; this must have length equal to the number of variables. This may also be a logical vector giving which variables to label. The default is to use the appropriate dimnames from bivars.

expand:

a number giving the factor by which to expand or contract the computed scale for the variables.

xlim:

the limits for the x-axis on the scale of the observations.

ylim:

the limits for the y-axis on the scale of the observations.

Graphical parameters may also be supplied as arguments to this function (see par).

Gabriel, K. R. and Odoroff, C. L. (1990). Biplots in biomedical research. Statistics in Medicine. 9 469-485.

Seber, G. A. F. (1984). Multivariate Observations. Wiley; New York.

longley.pc <- princomp(longley.x, cor=T)
biplot(longley.pc)
biplot(longley.pc, expand=1.1, var.axes=c(GNP=T,Arm=T,Unem=T))