[R] Obtaining the transformation matrix from the "persp" function without plotting

Erin Hodgess erinm.hodgess at gmail.com
Mon Jul 6 03:48:00 CEST 2015


Hello again.

I am looking at the "persp" function and would like to obtain the the
transformation matrix value that is output rather than the plot itself.  I
will use this matrix to pass into 3D plotting functions.

This is what I am doing:

library(rsm)
library(rgl)
tool.df <- read.table("tool.txt",header=TRUE)
tool.code <- coded.data(tool.df,x1~(speed-150)/25,x2~(angle-20)/5)
tool.rsm <- rsm(life ~ SO(x1,x2), data=tool.code)
summary(tool.rsm)

Note:  the tool.txt file is attached.

Update the model:

tool2.rsm = update(tool.rsm,  . ~ . + I(x1^2*x2) + I(x1*x2^2) +
I(x1^2*x2^2))

Now plot:

persp3d.rsm(tool2.rsm, x1~x2, bounds=list(x1=c(-1,1),x2=c(-1,1)),
zlim=c(-6,6),
        contour=TRUE)

Similarly, the persp3d.rsm.R file is attached.

As you can see, about half way in the persp3d.rsm function, I try to run a
regular persp function to get the transformation matrix:

transf2 = persp(dat$x, dat$y, dat$z, zlim = dat$zlim,
                  theta = theta, phi = phi, r = r, col = NA, border = NA,
                   box = FALSE, ...)

However, this produces a set of contour lines on a separate plot, rather
than the 3D plot.

I tried to used "plot=FALSE" in the persp function call, but that gave me a
warning message that the parameter did not exist.

This is from Russell Lenth's work.  Thanks to him for the rsm package.

Thank you for any suggestions.

Sincerely,
Erin



-- 
Erin Hodgess
Associate Professor
Department of Mathematical and Statistics
University of Houston - Downtown
mailto: erinm.hodgess at gmail.com
-------------- next part --------------
 speed angle life
    125    15   -2
    150    15   -3
    175    15    2
    125    15   -1
    150    15    0
    175    15    3
    125    20    0
    150    20    1
    175    20    4
    125    20    2
    150    20    3
    175    20    6
    125    25   -1
    150    25    5
    175    25    0
    125    25    0
    150    25    6
    175    25   -1


More information about the R-help mailing list