predict.cobs {cobs}

Predict method for COBS Fits

Description

Compute predicted values and simultaneous or pointwise confidence bounds for cobs objects.

Usage

## 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, ...)

Arguments

object

object of class cobs.

z

vector of grid points at which the fitted values are evaluated; defaults to an equally spaced grid with nz grid points between minz and maxz. Note that now z may lie outside of the knots interval which was not allowed originally.

deriv

scalar integer specifying (the order of) the derivative that should be computed.

minz

numeric needed if z is not specified; defaults to min(x) or the first knot if knots are given.

maxz

analogous to minz; defaults to max(x) or the last knot if knots are given.

nz

number of grid points in z if that is not given; defaults to 100.

interval

type of interval calculation, see below

level

confidence level

...

further arguments passed to and from methods.

Value

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.

Author(s)

Martin Maechler, based on He and Ng's code in cobs().

See Also

cobs the model fitting function.

Examples

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)