Prior to using functions X as predictors in a scalar-on-function regression, it is often
necessary to presmooth curves to remove measurement error or interpolate to a common grid. This
function creates a function to do this preprocessing depending on the method specified.
create.prep.func(
X,
argvals = seq(0, 1, length = ncol(X)),
method = c("fpca.sc", "fpca.face", "fpca.ssvd", "bspline", "interpolate"),
options = NULL
)a function that returns the preprocessed functional predictors, with arguments
The functional predictors to process
Indices of evaluation of newX
Any options needed to preprocess the predictor functions
an N by J=ncol(argvals) matrix of function evaluations
\(X_i(t_{i1}),., X_i(t_{iJ}); i=1,.,N.\) For FPCA-based processing methods, these functions are
used to define the eigen decomposition used to preprocess current and future data (for example, in
predict.pfr)
matrix (or vector) of indices of evaluations of \(X_i(t)\); i.e. a matrix with ith row \((t_{i1},.,t_{iJ})\)
character string indicating the preprocessing method. Options
are "fpca.sc", "fpca.face", "fpca.ssvd", "bspline",
and "interpolate". The first three use the corresponding existing function;
"bspline" uses an (unpenalized) cubic bspline smoother with nbasis basis
functions; "interpolate" uses linear interpolation.
list of options passed to the preprocessing method; as an example, options for fpca.sc
include pve, nbasis, and npc.
Jeff Goldsmith ajg2202@cumc.columbia.edu
pfr, fpca.sc, fpca.face, fpca.ssvd