pid (version 0.50)

contourPlot: 2-dimensional contour plot of a factorial design model

Description

Creates a two-dimensional contour plot of a factorial design model (linear least squares model). The two specified factors are plotted on the horizontal and vertical axes. If the model has more than two factors, then the remaining factors are set at their zero level. It is assumed the model is in coded units.

Usage

contourPlot(lsmodel, 
            xlab=attr(lsmodel$terms,'term.labels')[1], 
            ylab=attr(lsmodel$terms,'term.labels')[2], 
            main="Contour plot", 
            N=25, 
            xlim=c(-3.2, 3.2), 
            ylim=c(-3.2, 3.2),
            colour.function=terrain.colors)

Arguments

lsmodel

a linear model object (least squares model) created by the lm(…) function.

xlab

the label and variable used for horizontal axis in the bar plot, and must also match a variable in the lsmodel; the default will use the first available variable.

ylab

the label and variable used for vertical axis in the bar plot, and must also match a variable in the lsmodel; the default will use the second available variable.

main

label for the plot.

N

the resolution of the plot; higher values have greater resolution.

xlim

the limits of the horizontal axis (in coded units); the defaults indicate the model's effect outside the standard factorial range.

ylim

the limits of the vertical axis (in coded units); the defaults indicate the model's effect outside the standard factorial range.

colour.function

the function used to colour-code the contour plot; see rainbow for alternative colour schemes.

Value

Returns a ggplot2 object, which, by default, is shown before being returned by this function. The ggplot2 object may be further manipulated, if desired.

Details

Typical usage is to create a generic linear model with the lm(…) command, and supply that as the input to this function.

For example, a general design of experiments with 4 factors: A, B, C, and D can be built using lsmodel <- lm(y ~ A*B*C*D), and then various 2-D contour plots visualized with contourPlot(lsmodel, "A", "B"), contourPlot(lsmodel, "B", "C"), or whichever combination of factors are desired. The variables not being plotted are set at their zero (0) value.

Future versions of this function will allow specifying the inactive variable levels. In the interim, please see the rsm package (the rsm::contour(...) function) if this functionality is required.

References

Please see Chapter 5 of the following book: Kevin Dunn, 2010 to 2019, Process Improvement using Data, https://learnche.org/pid

Examples

# NOT RUN {
# 2-factor example
# ----------------
T <- c(-1, +1, -1, +1)  # centered and scaled temperature
S <- c(-1, -1, +1, +1)  # centered and scaled speed variable
y <- c(69, 60, 64, 53)  # conversion, is our response variable, y
doe.model <- lm(y ~ T + S + T * S)  # model with main effects, and interaction
contourPlot(doe.model)  

# 3-factor example
# ----------------
data(pollutant)
mod.full <- lm(y ~ C*T*S, data=pollutant) 
contourPlot(mod.full, N=50)               # high resolution plot
contourPlot(mod.full, xlab='C', ylab='S',
            main="Strong C:S interaction",
			colour.function=rainbow)
			
# Central composite design
P <- c(-1,   +1,  -1,  +1,     0, -1.41,     0, +1.41)
T <- c(-1,   -1,  +1,  +1, +1.41,     0, -1.41,     0)
y <- c(715, 713, 733, 725,   738,   717,   721,   710)
mod.CCD <- lm(y ~ P*T + I(P^2) + I(T^2))  
contourPlot(mod.CCD, 'P', 'T', xlim=c(-2.2, 2.2), ylim=c(-3,2))
# }