spatpcaCVWithSelectingK: Internal function: M-fold CV of SpatPCA with selecting K

In SpatPCA: Regularized Principal Component Analysis for Spatial Data

Description

Internal function: M-fold CV of SpatPCA with selecting K

Usage

spatpcaCVWithSelectingK(
  x,
  Y,
  M,
  tau1,
  tau2,
  gamma,
  shuffle_split,
  maxit,
  thr,
  l2
)

Arguments

x	Location matrix
Y	Data matrix
M	The number of folds for cross validation; default is 5.
tau1	Vector of a nonnegative smoothness parameter sequence. If NULL, 10 tau1 values in a range are used.
tau2	Vector of a nonnegative sparseness parameter sequence. If NULL, none of tau2 is used.
gamma	Vector of a nonnegative hyper parameter sequence for tuning eigenvalues. If NULL, 10 values in a range are used.
shuffle_split	Vector of indeces for random splitting Y into training and test sets
maxit	Maximum number of iterations. Default value is 100.
thr	Threshold for convergence. Default value is 10^{-4}.
l2	Vector of a nonnegative tuning parameter sequence for ADMM use

Value

A list of objects including

cv_result	A list of resultant objects produced by spatpcaCV
selected_K	Selected K based on CV.

SpatPCA documentation built on Jan. 31, 2021, 5:05 p.m.