Proximities between functional data

Approximates semi-metric distances for functional data of class fdata or fd.

Usage

semimetric.basis(
  fdata1,
  fdata2 = fdata1,
  nderiv = 0,
  type.basis1 = NULL,
  nbasis1 = NULL,
  type.basis2 = type.basis1,
  nbasis2 = NULL,
  ...
)

Arguments

fdata1: Functional data 1 or curve 1.
fdata2: Functional data 2 or curve 2.
nderiv: Order of derivation, used in deriv.fd
type.basis1: Type of Basis for fdata1.
nbasis1: Number of Basis for fdata1.
type.basis2: Type of Basis for fdata2.
nbasis2: Number of Basis for fdata2.
...: Further arguments passed to or from other methods.

Value

Returns a proximities matrix between functional data.

Details

Approximates semi-metric distances for functional data of two fd class objects. If functional data are not functional fd class, the semimetric.basis function creates a basis to represent the functional data, by default is used create.bspline.basis and the fdata class object is converted to fd class using the Data2fd.
The function calculates distances between the derivative of order nderiv of curves using deriv.fd function.

References

Ferraty, F. and Vieu, P. (2006). Nonparametric functional data analysis. Springer Series in Statistics, New York.

Examples

if (FALSE) { # \dontrun{
data(phoneme)
DATA1<-phoneme$learn[c(30:50,210:230)]
DATA2<-phoneme$test[231:250]
a1=semimetric.basis(DATA1,DATA2)
a2=semimetric.basis(DATA1,DATA2,type.basis1="fourier",
nbasis1=11, type.basis2="fourier",nbasis2=11)
fd1 <- fdata2fd(DATA1)
fd2 <- fdata2fd(DATA2)
a3=semimetric.basis(fd1,fd2)
a4=semimetric.basis(fd1,fd2,nderiv=1)
} # }