prcomp and princomp

Brian Ripley · 1998-08-24T15:35:17Z

> plot.prcomp xlab=paste("Principle component", x), > ylab=paste("Principle component", y), ...) { > if (is.null(obj$x)) > stop("Rotated x has not be calculated by prcomp. Use retx=T in prcomp.") > plot(obj$x[,c(x,y)], main=main, xlab=xlab, ylab=ylab, ...) > } > plot.princomp xlab=paste("Principle component", x), > ylab=paste("Principle component", y), ...) { > plot(obj$scores[,c

2 messages · Brian Ripley, Paul Gilbert

Original

Brian Ripley

Mon, Aug 24, 1998 8:35 AM #

Ouch. Enough of my students think that it is the spelling already!
This is _not_ what is meant by a scree plot, which plots the singular
values or eigenvalues.

Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Paul Gilbert

Mon, Aug 24, 1998 9:50 AM #

Apologies. I thought I had eradicated this.

My mistake here was in the title main="Scree Plot", which in the end was not
what I decided to plot. At one point I thought the default plot method should be
the scree plot, but later decided the default plot method should plot the
principal components.  I noticed that in princomp-add.R you have

plot.princomp <- function(...) screeplot(...)

so perhaps you have different thoughts on this? I would propose having the
default plot the principal components (with the correct title) and use your
function scree, but not as the default.

Paul Gilbert


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._