smooth.construct.fpc.smooth.spec {refund}

Basis constructor for FPC terms

Description

Basis constructor for FPC terms

Usage

## S3 method for class 'fpc.smooth.spec'
smooth.construct(object, data, knots)

Arguments

object

a fpc.smooth.spec object, usually generated by a term s(x, bs="fpc"); see Details.

data

a list containing the data (including any by variable) required by this term, with names corresponding to object$term (and object$by). Only the first element of this list is used.

knots

not used, but required by the generic smooth.construct.

Details

object must contain an xt element. This is a list that can contain the following elements:

X

(required) matrix of functional predictors

method

(required) the method of finding principal components; options include "svd" (unconstrained), "fpca.sc", "fpca.face", or "fpca.ssvd"

npc

(optional) the number of PC's to retain

pve

(only needed if npc not supplied) the percent variance explained used to determine npc

penalize

(required) if FALSE, the smoothing parameter is set to 0

bs

the basis class used to pre-smooth X; default is "ps"

Any additional options for the pre-smoothing basis (e.g. k, m, etc.) can be supplied in the corresponding elements of object. See s for a full list of options.

Value

An object of class "fpc.smooth". In addtional to the elements listed in smooth.construct, the object will contain

sm

the smooth that is fit in order to generate the basis matrix over object$term

V.A

the matrix of principal components

Author(s)

Jonathan Gellar [e-mail]

References

Reiss, P. T., and Ogden, R. T. (2007). Functional principal component regression and functional partial least squares. Journal of the American Statistical Association, 102, 984-996.

See Also

fpcr