contourPlot3: function to create a ternary plot (contour plot) for an...

In qualityTools: Statistical Methods for Quality Science

Description

This function creates a ternary plot (contour plot) for mixture designs (i.e. object of class mixDesign)

Usage

contourPlot3(x, y, z, response, data = NULL, main, xlab, ylab, zlab, border, 
             form = "linear", col = 1, col.text, cex.axis, axes = TRUE, 
             steps, factors)

Arguments

x	factor 1 of the mixDesign object.
y	factor 2 of the mixDesign object.
z	factor 3 of the mixDesign object.
response	the response of the mixDesign object.
data	the mixDesign object from which x,y,z and the response are taken.
main	giving an overall title for the plot: see title.
xlab	giving a title for the x axis : see title.
ylab	giving a title for the y axis : see title.
zlab	giving a title for the z axis : see title.
border	numeric or character (for example “red”) value specifying the color of the surroundimg the ternary plot. By default border is set to “white”
form	a character string or a formula with the syntax “y ~ A + B +C”. If form is a character string it has to be one out of the following: “linear” “quadratic” “fullCubic” “specialCubic” How the form influences the output is described in the reference listed below. By default form is set as “linear”.
col	a predefined (1, 2, 3 or 4) or self defined colorRampPalette.
col.text	a numerical value or a character string (like “red”) giving the color oft the inscription of the axes. The default value for col.text is ‘1’.
cex.axis	numerical value giving the size of the inscription of the axes. The default value cex.axis is ‘1’.
axes	logical value specifying whether the axes should be plotted or not. By default axes is set to ‘TRUE’.
steps	resolution of the plot (number of rows for the square matrix), also number of grid points per factor. By default steps = 25.
factors	list of factors for categorizing with setting in case there are more than 3 factors (not yet implemented).

Value

contourPlot3 returns an invisible matrix containing the response values as NA's and numerics.

Note

This method is still under construction!
For an example in context which shows the usage of the function contourPlot3() to an object of class mixDesign, please read the vignette for the package qualityTools at http://www.r-qualitytools.org/html/Improve.html.

Author(s)

Thomas Roth [e-mail]

References

MYERS, Raymond H.; MONTGOMERY, Douglas C.; ANDERSON-COOK, Christine M.: Response Surface Methodology. New York: WILEY ,2009

See Also

wirePlot3
mixDesign
colorRampPalette
http://www.r-qualitytools.org/html/Improve.html

Examples

#yarn example p.564 Response Surface Methodology
mdo = mixDesign(3,2, center = FALSE, axial = FALSE, randomize = FALSE, 
                replicates  = c(1,1,2,3))
names(mdo) = c("polyethylene", "polystyrene", "polypropylene")
units(mdo) = "percent"
elongation = c(11.0, 12.4, 15.0, 14.8, 16.1, 17.7, 16.4, 16.6, 8.8, 10.0, 10.0, 
               9.7, 11.8, 16.8, 16.0)
response(mdo) = elongation

dev.new(14,14); par(mfrow = c(2,2))
contourPlot3(A, B, C, elongation, data = mdo, form = "linear")
contourPlot3(A, B, C, elongation, data = mdo, form = "quadratic", col = 2)
contourPlot3(A, B, C, elongation, data = mdo, 
             form = "elongation ~ I(A^2) - B:A + I(C^2)", col = 3)
contourPlot3(A, B, C, elongation, data = mdo, form = "quadratic", 
             col = colorRampPalette(c("yellow", "white", "red")))

Example output

Loading required package: Rsolnp
Loading required package: MASS

Attaching package: 'qualityTools'

The following object is masked from 'package:stats':

    sigma

Warning messages:
1: In `[<-`(`*tmp*`, i, value = <S4 object of class "doeFactor">) :
  implicit list embedding of S4 objects is deprecated
2: In `[<-`(`*tmp*`, i, value = <S4 object of class "doeFactor">) :
  implicit list embedding of S4 objects is deprecated
3: In `[<-`(`*tmp*`, i, value = <S4 object of class "doeFactor">) :
  implicit list embedding of S4 objects is deprecated
[1] "elongation"

qualityTools documentation built on May 2, 2019, 10:21 a.m.