plot.design
Plot Univariate Effects of a Design or Model
Plot univariate effects of one or more factor
s,
typically for a designed experiment as analyzed by aov()
.
- Keywords
- hplot
Usage
plot.design(x, y = NULL, fun = mean, data = NULL, ..., ylim = NULL, xlab = "Factors", ylab = NULL, main = NULL, ask = NULL, xaxt = par("xaxt"), axes = TRUE, xtick = FALSE)
Arguments
- x
- either a data frame containing the design factors and
optionally the response, or a
formula
orterms
object. - y
- the response, if not given in x.
- fun
- a function (or name of one) to be applied to each subset. It must return one number for a numeric (vector) input.
- data
- data frame containing the variables referenced by
x
when that is formula-like. - ...
- graphical parameters such as
col
, seepar
. - ylim
- range of y values, as in
plot.default
. - xlab
- x axis label, see
title
. - ylab
- y axis label with a smart default.
- main
- main title, see
title
. - ask
- logical indicating if the user should be asked before a new page is started -- in the case of multiple y's.
- xaxt
- character giving the type of x axis.
- axes
- logical indicating if axes should be drawn.
- xtick
- logical indicating if ticks (one per factor) should be drawn on the x axis.
Details
The supplied function will be called once for each level of each
factor in the design and the plot will show these summary values. The
levels of a particular factor are shown along a vertical line, and the
overall value of fun()
for the response is drawn as a
horizontal line.
Note
A big effort was taken to make this closely compatible to the S
version. However, col
(and fg
) specifications have
different effects.
In S this was a method of the plot
generic function for
design
objects.
References
Chambers, J. M. and Hastie, T. J. eds (1992) Statistical Models in S. Chapman & Hall, London, the white book, pp. 546--7 (and 163--4).
Freeny, A. E. and Landwehr, J. M. (1990) Displays for data from large designed experiments; Computer Science and Statistics: Proc.\ 22nd Symp\. Interface, 117--126, Springer Verlag.
See Also
interaction.plot
for a standard graphic
of designed experiments.
Examples
library(graphics)
require(stats)
plot.design(warpbreaks) # automatic for data frame with one numeric var.
Form <- breaks ~ wool + tension
summary(fm1 <- aov(Form, data = warpbreaks))
plot.design( Form, data = warpbreaks, col = 2) # same as above
## More than one y :
utils::str(esoph)
plot.design(esoph) ## two plots; if interactive you are "ask"ed
## or rather, compare mean and median:
op <- par(mfcol = 1:2)
plot.design(ncases/ncontrols ~ ., data = esoph, ylim = c(0, 0.8))
plot.design(ncases/ncontrols ~ ., data = esoph, ylim = c(0, 0.8),
fun = median)
par(op)