eigenf_plot: Plot of the vertical modes

In EPauthenet/fda.oce: Functional Data Analysis of oceanographic profiles.

Description

This function plots the effect of each vertical mode (i. e. eigenfunction) on the mean profile.

Usage

eigenf_plot(pca, te, sign = 1)

Arguments

pca	the list produced by the function fpca.
te	Which PC to plot.
sign	The PCs are invariant by their sign, so the choice of the sign depends on the feature to represent. The parameter sign can contain a vector of 1 or -1 to inverse the sign of the PCs if wanted. sign can also be used as a factor to increase the effect of eigenfunctions and see better the small variations.

Value

Plot of the effects of the vertical mode te on the mean profiles. The curves show the mean profile (solid) and the effects of adding (red) and subtracting (blue) eigenfunctions. The percentages next to the header titles are the amount of variance explained by the mode displayed. The percentages in the horizontal axis label are the variance contained by each variable (T and S) on the mode displayed. sign can be used as a factor to enhance the deformation of + and - curves.

Author(s)

Etienne Pauthenet <[e-mail]>, David Nerini <[e-mail]>, Fabien Roquet <[e-mail]>

References

Pauthenet et al. (2017) A linear decomposition of the Southern Ocean thermohaline structure. Journal of Physical Oceanography, http://dx.doi.org/10.1175/JPO-D-16-0083.1

Ramsay, J. O., and B. W. Silverman, 2005: Functional Data Analysis. 2nd Edition Springer, 426 pp., Isbn : 038740080X.

See Also

bspl for bsplines fit on T-S profiles, fpca for functional principal component analysis of T-S profiles, proj for computing Principal Components, reco for reconstructing profiles with less modes.

EPauthenet/fda.oce documentation built on Feb. 1, 2021, 4:34 p.m.