pcaadd: Principal component analysis with reference spectra

In cbmodels: Collection of "combined" models: PCA-LDA, PLS-LDA, etc.

Description

A PCA model is calculated after a multiple of the reference matrix is added to the data matrix.

Usage

  ## S4 method for signature 'matrix,matrix'
pcaadd(x, reference, ...,
    ref.factor = 10, refcomps = seq_len(nrow(reference)))

  ## S4 method for signature 'hyperSpec,hyperSpec'
pcaadd(x, reference, ...)

  ## S4 method for signature 'hyperSpec,matrix'
pcaadd(x, reference, ...)

  ## S4 method for signature 'matrix,hyperSpec'
pcaadd(x, reference, ...)

  ## S3 method for class 'pcaadd'
 predict(object, newdata, ...)

Arguments

x	data matrix
reference	reference data points
...	further arguments are handed to prcomp, but center is always FALSE.
ref.factor	reference is multiplied by ref.factor * max (abs (range (x))) / max (abs (range (reference))) * nrow (x) / nrow (reference) before the PCA. This forces the first directions towards reference.
refcomps	the principal components that are attributed to the reference
object	pcaadd model
newdata	matrix with new observations.

Value

object of class "pcaadd"

cbmodels documentation built on May 31, 2017, 2:11 a.m.