R/eval.fd.R

In fda: Functional Data Analysis

predict.fdSmooth <- function(object, newdata=NULL, Lfdobj=0,
                             returnMatrix=FALSE, ...){
  if(is.null(newdata)){
    newdata <- object$argvals
  }
  eval.fd(newdata, object$fd, Lfdobj, returnMatrix=returnMatrix)
}

fitted.fdSmooth <- function(object, returnMatrix=FALSE, ...){
  newdata <- object$argvals
  eval.fd(newdata, object$fd, 0, returnMatrix=returnMatrix)
}

residuals.fdSmooth <- function(object, returnMatrix=FALSE, ...){
  newdata <- object$argvals
  pred <- eval.fd(newdata, object$fd, 0, returnMatrix=returnMatrix)
  object$y-pred
}

predict.fdPar <- function(object, newdata=NULL, Lfdobj=0,
                          returnMatrix=FALSE, ...){
  predict.fd(object$fd, newdata, Lfdobj,
             returnMatrix=returnMatrix, ...)
}

predict.fd <- function(object, newdata=NULL, Lfdobj=0,
                       returnMatrix=FALSE, ...){
  if(is.null(newdata)){
    basis <- object$basis
    type <- basis$type
    if(length(type) != 1)
      stop('length(object$type) must be 1;  is ',
           length(type) )
    newdata <- {
      if(type=='bspline')
        unique(knots(basis, interior=FALSE))
      else basis$rangeval
    }
  }
  eval.fd(newdata, object, Lfdobj, returnMatrix=returnMatrix)
}

#  ----------------------------------------------------------------------------

eval.fd <- function(evalarg, fdobj, Lfdobj=0, returnMatrix=FALSE) {

#  EVAL_FD evaluates a functional data observation at argument
#  values EVALARG.
#
#  LFDOBJ is a functional data object defining the order m
#  HOMOGENEOUS linear differential operator of the form
#  Lx(t) = w_0(t) x(t) + ... + w_{m-1}(t) D^{m-1}x(t) +
#          \exp[w_m(t)] D^m x(t)
#
#  Arguments:
#  EVALARG ... A vector of values at which all functions are to
#              evaluated.
#  FDOBJ   ... Functional data object
#  LFDOBJ  ... A linear differential operator object
#              applied to the functions before they are evaluated.
#  RETURNMATRIX ... If False, a matrix in sparse storage model can be returned
#               from a call to function BsplineS.  See this function for
#               enabling this option.

#  Note that the first two arguments may be interchanged.

#  Returns:  An array of function values corresponding to the evaluation
#              arguments in EVALARG

#  Last modified Oct 19, 2012 by Spencer Graves

#  Check LFDOBJ

  Lfdobj <- int2Lfd(Lfdobj)

#  Exchange the first two arguments if the first is an FD object
#    and the second numeric

  if (inherits(fdobj, "numeric") && inherits(evalarg, "fd")) {
      temp    <- fdobj
      fdobj   <- evalarg
      evalarg <- temp
  }

#  check EVALARG

#  if (!(is.numeric(evalarg))) stop("Argument EVALARG is not numeric.")
  Evalarg <- evalarg
  if(!is.numeric(evalarg)){
    op <- options(warn=-1)
    evalarg <- as.numeric(Evalarg)
    options(op)
    nNA <- sum(is.na(evalarg))
    if(nNA>0)
      stop('as.numeric(evalarg) contains ', nNA,
           ' NA', c('', 's')[1+(nNA>1)],
           ';  class(evalarg) = ', class(Evalarg))
  }

  evaldim <- dim(evalarg)
  if (!(length(evaldim) < 3))
      stop("Argument 'evalarg' is not a vector or a matrix.")

#  check FDOBJ

  if (!(inherits(fdobj, "fd")))
      stop("Argument FD is not a functional data object.")

#  Extract information about the basis

  basisobj <- fdobj$basis
  nbasis   <- basisobj$nbasis
  rangeval <- basisobj$rangeval
  onerow   <- rep(1,nbasis)

  temp <- c(evalarg)
  temp <- temp[!(is.na(temp))]
  EPS  <- 5*.Machine$double.eps
  if (min(temp) < rangeval[1]-EPS || max(temp) > rangeval[2]+EPS) {
    warning(paste("Values in argument 'evalarg' are outside ",
                  "of permitted range and will be ignored."))
    print(c(rangeval[1]-min(temp), max(temp) - rangeval[2]))
  }

#  get maximum number of evaluation values

  if (is.vector(evalarg)) {
      n <- length(evalarg)
  } else {
      n <- evaldim[1]
  }

#  Set up coefficient array for FD

  coef  <- fdobj$coefs
  coefd <- dim(coef)
  ndim  <- length(coefd)
  if (ndim <= 1) nrep <- 1 else nrep <- coefd[2]
  if (ndim <= 2) nvar <- 1 else nvar <- coefd[3]

# check coef is conformable with evalarg

  if(length(evaldim)>1){
    if(evaldim[2]==1){
      evalarg <- c(evalarg)
    } else {
      if(evaldim[2] != coefd[2]){
        stop('evalarg has ', evaldim[2], ' columns;  does not match ',
             ndim[2], ' = number of columns of ffdobj$coefs')
      }
    }
  }

#  Set up array for function values

  if (ndim <= 2) {
      evalarray <- matrix(0,n,nrep)
  } else evalarray <- array(0,c(n,nrep,nvar))
  if (ndim == 2) dimnames(evalarray) <- list(NULL,dimnames(coef)[[2]])
  if (ndim == 3)
      dimnames(evalarray) <- list(NULL,dimnames(coef)[[2]],
                                  dimnames(coef)[[3]])

#  Case where EVALARG is a vector of values to be used for all curves

  if (is.vector(evalarg)) {

    evalarg[evalarg < rangeval[1]-1e-10] <- NA
    evalarg[evalarg > rangeval[2]+1e-10] <- NA
    basismat <- eval.basis(evalarg, basisobj, Lfdobj, returnMatrix)

    #  evaluate the functions at arguments in EVALARG

    if (ndim <= 2) {
      evalarray <- basismat %*% coef
#     needed because dimnames may malfunction with Matrix basismat
      dimnames(evalarray) <- list(rownames(basismat), colnames(coef))
    } else {
       evalarray <- array(0,c(n,nrep,nvar))
       for (ivar in 1:nvar) evalarray[,,ivar] <- basismat %*% coef[,,ivar]
    }

  } else {

  #  case of evaluation values varying from curve to curve

      for (i in 1:nrep) {
        evalargi <- evalarg[,i]
        if (all(is.na(evalargi))) stop(
            paste("All values are NA for replication",i))

        index    <- !(is.na(evalargi) | evalargi < rangeval[1] |
                                    evalargi > rangeval[2])
        evalargi <- evalargi[index]
        basismat <- eval.basis(evalargi, basisobj, Lfdobj, returnMatrix)

       #  evaluate the functions at arguments in EVALARG

        if (ndim == 2) {
            evalarray[  index, i] <- as.vector(basismat %*% coef[,i])
            evalarray[!(index),i] <- NA
        }
        if (ndim == 3) {
            for (ivar in 1:nvar) {
                evalarray[   index,i,ivar] <-
                    as.vector(basismat %*% coef[,i,ivar])
                evalarray[!(index),i,ivar] <- NA
            }
        }
    }
  }

  if((length(dim(evalarray))==2) && !returnMatrix) {
      return(as.matrix(evalarray))
  } else return(evalarray)
}