color.plot.loess {mining} | R Documentation |
color.plot.loess(object,data,res=50,fill=F,add=fill,clip=T,...)
object |
a loess object. |
data |
data to use instead of model.frame(object) . |
res |
resolution of the sampling grid in each direction. |
fill |
passed to contour.plot . |
add |
If TRUE , the contours are added to an existing
plot. Otherwise a new plot is created. |
clip |
a polygon over which the surface is to be defined.
Possible values are FALSE (no clipping), TRUE
(clip to the convex hull of the data), or a
matrix with two columns specifying (x,y) coordinates. |
... |
extra arguments to color.plot.data.frame |
.
The regression surface is evaluated at all points on a grid,
clipped values are set to NA
, and contour.plot
is
used to plot the contours.
If add=FALSE
, the data is plotted on top using
color.plot.data.frame
.
A plot is produced.
Tom Minka
data(Housing) fit = loess(Price ~ Rooms + Low.Status, Housing) color.plot(fit)