margin.plot.lm: Margin Plots of Response for Pairs of Factors in Linear Model

In byandell/pda: Practial Data Analysis for Designed Experiments

Usage

## S3 method for class 'lm'
margin.plot(object, reduced, data, factors, lsm, margin,
   error = "full", bar.plot = "lsd",
   xlab, ylab, ...)

Arguments

object	full model object from linear model fit
reduced	reduced model object, for computation of marginal means or effects(default=object)
data	data frame in which to interpret the variables named in the full and reduced objects (determined from reduced fit if omitted)
factors	Character string of length 2 with names of x.factor and trace.factoras found in fit and data (default is first 2 factors in fit).
lsm	Least squares means for plotting (default is fit from full model).
margin	Least squares marginal means for horizontal axis(default is fit from reduced model rounded to proper precision).
error	model to use for error calculation ("full" or "reduced").
bar.plot	error bars:Fisher's LSD ("lsd"=default)or confidence intervals ("ci")or none ("none").
xlab,ylab	plot parameters
...	optional specification of other parameters to be passed to plotting routines.

Value

a plot will be created showing the requested function of responses for each level of the x.factor at each level of the trace.factor. By default, lines for each value of the trace.factor are drawn in different styles so that they may be more easily distinguished.

See Also

lsd.plot, margin.plot, tukey.plot.

Examples

data(ToothGrowth)
ToothGrowth$dose = ordered(ToothGrowth$dose)
reduced <- lm(len~dose+supp, ToothGrowth)
full <- update(reduced,len~dose*supp)
margin.plot.lm(full,reduced,ToothGrowth)

byandell/pda documentation built on May 13, 2019, 9:27 a.m.