SelectK: Selects number of functional principal components for given...

In fdapace: Functional Data Analysis and Empirical Dynamics

Selects number of functional principal components for given FPCA output and selection criteria

Description

Selects number of functional principal components for given FPCA output and selection criteria

Usage

SelectK(fpcaObj, criterion = "FVE", FVEthreshold = 0.95, Ly = NULL, Lt = NULL)

Arguments

fpcaObj	A list containing FPCA related objects returned by MakeFPCAResults().
criterion	A string or positive integer specifying selection criterion for the number of functional principal components. Available options: 'FVE', 'AIC', 'BIC', or the specified number of components - default: 'FVE' For explanations of these criteria, see Yao et al (2005, JASA)
FVEthreshold	A threshold fraction to be specified by the user when using "FVE" as selection criterion: (0,1] - default: NULL
Ly	A list of n vectors containing the observed values for each individual - default: NULL
Lt	A list of n vectors containing the observation time points for each individual corresponding to Ly - default: NULL

Value

A list including the following two fields:

K	An integer indicating the selected number of components based on given criterion.
criterion	The calculated criterion value for the selected number of components, i.e. FVE, AIC or BIC value, NULL for fixedK criterion.

fdapace documentation built on July 3, 2024, 5:08 p.m.