predict.cobs | R Documentation |
Compute predicted values and simultaneous or pointwise confidence
bounds for cobs
objects.
## S3 method for class 'cobs'
predict(object, z, deriv = 0L,
minz = knots[1], maxz = knots[nknots], nz = 100,
interval = c("none", "confidence", "simultaneous", "both"),
level = 0.95, ...)
object |
object of class |
z |
vector of grid points at which the fitted values are
evaluated; defaults to an equally spaced grid with |
deriv |
scalar integer specifying (the order of) the derivative that should be computed. |
minz |
numeric needed if |
maxz |
analogous to |
nz |
number of grid points in |
interval |
type of interval calculation, see below |
level |
confidence level |
... |
further arguments passed to and from methods. |
a matrix of predictions and bounds if interval
is set (not
"none"). The columns are named z
, fit
, further
cb.lo
and cb.up
for the simultaneous
confidence
band, and ci.lo
and ci.up
the pointwise
confidence
intervals according to specified level
.
If z
has been specified, it is unchanged in the result.
Martin Maechler, based on He and Ng's code in cobs()
.
cobs
the model fitting function.
example(cobs) # continuing :
(pRbs <- predict(Rbs))
#str(pSbs <- predict(Sbs, xx, interval = "both"))
str(pSbs <- predict(Sbs, xx, interval = "none"))
plot(x,y, xlim = range(xx), ylim = range(y, pSbs[,2], finite = TRUE),
main = "COBS Median smoothing spline, automatical lambda")
lines(pSbs, col = "red")
lines(spline(x,f.true), col = "gray40")
#matlines(pSbs[,1], pSbs[,-(1:2)],
# col= rep(c("green","blue"),c(2,2)), lty=2)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.