screeplot.MFPCAfit function

Screeplot for Multivariate Functional Principal Component Analysis

This function plots the proportion of variance explained by the leading eigenvalues in an MFPCA against the number of the principal component.

## S3 method for class 'MFPCAfit'
screeplot(
  x,
  npcs = min(10, length(x$values)),
  type = "lines",
  ylim = NULL,
  main = deparse(substitute(x)),
  ...
)

Arguments

• x: An object of class MFPCAfit, typically returned by a call to MFPCA.
• npcs: The number of eigenvalued to be plotted. Defaults to all eigenvalues if their number is less or equal to 10, otherwise show only the leading first 10 eigenvalues.
• type: The type of screeplot to be plotted. Can be either "lines" or "barplot". Defaults to "lines".
• ylim: The limits for the y axis. Can be passed either as a vector of length 2 or as NULL (default). In the second case, ylim is set to (0,max(pve)), with pve the proportion of variance explained by the principal components to be plotted.
• main: The title of the plot. Defaults to the variable name of x.
• ...: Other graphic parameters passed to plot.default (for type = "lines") or barplot (for type = "barplot").

Returns

A screeplot, showing the decrease of the principal component score.

Examples

# Simulate multivariate functional data on one-dimensonal domains
# and calculate MFPCA (cf. MFPCA help)
set.seed(1)
# simulate data (one-dimensional domains)
sim <-  simMultiFunData(type = "split", argvals = list(seq(0,1,0.01), seq(-0.5,0.5,0.02)),
                       M = 5, eFunType = "Poly", eValType = "linear", N = 100)
# MFPCA based on univariate FPCA
PCA <- MFPCA(sim$simData, M = 5, uniExpansions = list(list(type = "uFPCA"),
                                                     list(type = "uFPCA")))

# screeplot
screeplot(PCA) # default options
screeplot(PCA, npcs = 3, type = "barplot", main= "Screeplot")

See Also

MFPCA, screeplot