incrementalPCA: incrementalPCA

In jdestefani/ExtendedDFML: Extended Dynamic Factor Machine Learner

Description

Wrapper function for an iterative PCA method (onlinePCA::incRpca)

Usage

incrementalPCA(
  X,
  xbar = NULL,
  base_pca_values,
  base_pca_vectors,
  pca_type = INCREMENTAL_PCA_METHODS,
  forgetting_factor = NA,
  components = 2,
  start_value,
  end_value
)

Arguments

X	n x N matrix containing the N time series as columns, each one of length n time steps
xbar	1 x N matrix containing the means of the N input time series
base_pca_values	Matrix containing the PCA eigenvalues to use for the initialization
base_pca_vectors	Matrix containing the PCA eigenvectors to use for the initialization
pca_type	Incremental PCA method to employ - String among those defined in INCREMENTAL_PCA_METHODS
forgetting_factor	Forgetting factor to employ in the incremental PCA - Numeric
components	Desired number of PCA components - Numeric
start_value	Index of the first value in the dataset to be used for incremental PCA - Numeric
end_value	Index of the last value in the dataset to be used for incremental PCA - Numeric

Value

List containing:

pca - List containing:

• values: Matrix containing the eigenvalues associated to PCA

• vectors: Matrix containing the eigenvectors associated to PCA as columns

• x_bar: 1 x N matrix containing the means of the N input time series

jdestefani/ExtendedDFML documentation built on Dec. 20, 2021, 10:04 p.m.