smooth.sparse.mean: Smooth the mean function of sparse data

In fda: Functional Data Analysis

Smooth the mean function of sparse data

Description

Do a smoothing of the mean function for sparse data that is either given as a list or as a matrix with NAs. The smooth is done by basis expansion with the functional basis "type"; if !(lambda == 0) then the second derivative is penalized (int2Lfd(2)).

Usage

smooth.sparse.mean(data, time ,rng = c(0, 1), type = "", nbasis = NULL, 
                  knots = NULL, norder = NULL, lambda = NULL)

Arguments

data	a matrix object or list – If the set is supplied as a matrix object, the rows must correspond to argument values and columns to replications, and it will be assumed that there is only one variable per observation. If y is a three-dimensional array, the first dimension corresponds to argument values, the second to replications, and the third to variables within replications. – If it is a list, each element must be a matrix object, the rows correspond to argument values per individual. First column corresponds to time points and followins columns to argument values per variable.
time	Array with time points where data was taken. length(time) == ncol(data)
rng	an array of length 2 containing the lower and upper boundaries for the rangeval of argument values
type	Type of basisfd for smoothing the mean estimate function. "bspline", "fourier", "exp", "const" or "mon"
nbasis	An integer variable specifying the number of basis functions
knots	a vector specifying the break points if type == "bspline"
norder	an integer specifying the order of b-splines if type == "bspline"
lambda	a nonnegative real number specifying the amount of smoothing to be applied to the estimated functional parameter

Value

a functional data object containing a smooth of the mean.

fda documentation built on May 31, 2023, 9:19 p.m.