| KernSmooth-internal {KernSmooth} | R Documentation |
Internal KernSmooth functions
blkest(x, y, Nval, q)
cpblock(X, Y, Nmax, q)
linbin(X, gpoints, truncate=TRUE)
linbin2D(X, gpoints1, gpoints2)
rlbin(X, Y, gpoints, truncate=TRUE)
sdiag(x, drv=0, degree=1, kernel="normal",
bandwidth, gridsize=401, bwdisc=25, range.x,
binned=FALSE, truncate=TRUE)
sstdiag(x, drv=0, degree=1, kernel="normal",
bandwidth, gridsize=401, bwdisc=25, range.x,
binned=FALSE, truncate=TRUE)
These are not to be called by the user.
blkest is used to obtain preliminary estimates of
quantities required for the ``direct plug-in''
regression bandwidth selector based on
blocked qth degree polynomial fits.
cpblock chooses the number of blocks for the preliminary
step of a plug-in rule using Mallows' C_p.
linbin applies linear binning to a univariate data set.
linbin2D applies linear binning to a bivariate data set.
rlbin applies linear binning to a regression data set.
sdiag computes the binned diagonal entries of a smoother
matrix for local polynomial kernel regression.
sstdiag computes the binned diagonal entries of SS^T
where S is a smoother matrix for local polynomial
kernel regression.