scatter3dPETER: 3D plot of a Regression Model

Description

This function makes a 3D plot of the data and the regression function. The user has the choice between different methods to calculate the coefficients for the regression model.

Usage

scatter3dPETER(x, y, z, xlab = deparse(substitute(x)),
ylab = deparse(substitute(y)), zlab = deparse(substitute(z)),
revolutions = 0, bg.col = c("white", "black"),
axis.col = if (bg.col == "white") "black" else "white",
surface.col = c("blue", "green", "orange", "magenta", "cyan", "red",
"yellow", "gray"), neg.res.col = "red",
pos.res.col = "green", point.col = "yellow", text.col = axis.col,
grid.col = if (bg.col == "white") "black" else "gray",
fogtype = c("exp2", "linear", "exp", "none"),
residuals = (length(fit) == 1), surface = TRUE, grid = TRUE,
grid.lines = 26, df.smooth = NULL, df.additive = NULL, sphere.size = 1,
threshold = 0.01, speed = 1, fov = 60, fit = "linear", groups = NULL,
parallel = TRUE, model.summary = FALSE)

Arguments

x, y, z

the coordinates for the points

xlab, ylab, zlab

the labels for the axis

revolutions

if the plot should be viewed from different angles

bg.col, axis.col, surface.col, point.col, text.col, grid.col

define the colour for the background, axis,...

pos.res.col, neg.res.col

colour for positive and negativ residuals

fogtype

describes the fogtype, see rgl.bg

residuals

if the residuals should be plotted

surface

if the regression function should be plotted or just the points

grid

if TRUE, the grid is plotted

grid.lines

number of lines in the grid

df.smooth

if fit=smooth, the number of degrees of freedom

df.additive

if fit=additive, the number of degrees of freedom

sphere.size

a value for calibrating the size of the sphere

threshold

the minimum size of the sphere, if the size is smaller than the threshold a point is plotted

speed

if revolutions>0, how fast you make a 360 degree turn

fov

field-of-view angle, see rgl.viewpoint

fit

which method should be used for the model; "linear", "quadratic", "smooth" or "additive"

groups

define groups for the points

parallel

if groups is not NULL, a parallel shift in the model is made

model.summary

if the summary should be returned

Details

The user can choose between a linear, quadratic, smoothed or additve model to calculate the coefficients.

References

C. Reimann, P. Filzmoser, R.G. Garrett, and R. Dutter: Statistical Data Analysis Explained. Applied Environmental Statistics with R. John Wiley and Sons, Chichester, 2008.

Examples

Run this code

#required library
#require(IPSUR)
data(chorizon)
lit=1
# This example needs additional libraries:
#scatter3dPETER(x=log10(chorizon[chorizon$LITO==lit,"Cr"]), 
#               z=log10(chorizon[chorizon$LITO==lit,"Cr_INAA"]),
#               y=log10(chorizon[chorizon$LITO==lit,"Co"]),
#               xlab="",ylab="",zlab="",
#               neg.res.col=gray(0.6), pos.res.col=gray(0.1), point.col=1, fov=30,
#               surface.col="black",grid.col="gray",sphere.size=0.8)