Cross-Validation of Classical PCA Results for a Spectra Object

This function carries out classical PCA on the data in a Spectra object using a cross-validation method. A simple re-write of Peter Filzmoser's pcaCV method with some small plotting changes.

cv_pcaSpectra(
  spectra,
  pcs,
  choice = "noscale",
  repl = 50,
  segments = 4,
  segment.type = c("random", "consecutive", "interleaved"),
  length.seg,
  trace = FALSE,
  ...
)

Arguments

spectra: An object of S3 class Spectra().
pcs: As per pcaCV where it is called amax; an integer giving the number of PC scores to include.
choice: A character string indicating the choice of scaling. One of c("noscale", "autoscale", "Pareto").
repl: As per pcaCV; the number of replicates to perform.
segments: As per pcaCV.
segment.type: As per pcaCV.
length.seg: As per pcaCV.
trace: As per pcaCV.
...: Parameters to be passed to the plotting routines.

Value

Invisibly, a list as described in pcaCV. Side effect is a plot.

References

K. Varmuza and P. Filzmoser Introduction to Multivariate Statistical Analysis in Chemometrics, CRC Press, 2009.

Author

Bryan A. Hanson, DePauw University. Derived from pcaCV.

Examples

# You need to install package "pls" for this example
if (requireNamespace("pls", quietly = TRUE)) {
  data(SrE.IR)
  pca <- cv_pcaSpectra(SrE.IR, pcs = 5)
}