incrementalDimensionalityReduction: incrementalDimensionalityReduction

In jdestefani/ExtendedDFML: Extended Dynamic Factor Machine Learner

Description

Function implementing the dimensionality reduction module in the DFML architecture. The function implements an incremental dimensionality reduction model with a rolling approach. A percentage of X, defined by init_sample_percentage is used to initialize the dimensionality reduction technique. The remaining part of X is used to incrementally fit the dimensionality reduction model

Usage

incrementalDimensionalityReduction(
  X,
  init_sample_percentage,
  k,
  family = DIMENSIONALITY_METHODS,
  params = list(model = NULL)
)

Arguments

X	n x N matrix containing the N time series as columns, each one of length n time steps
init_sample_percentage	Numeric value indicating the percentage of X to be used for model fit - Numeric
k	Number of desired factors as output - Numeric
family	Family of methods to use - String among those defined in DIMENSIONALITY_METHODS
params	Parameters to be passed to the function - List An (optional) pre-existing model is passed through the \code{parameters$model}. If NULL it will be computed by the function.

Value

List containing:

• Z: n x k matrix contaning the k factor generated by the dimensionality reduction module

• time_dim: Computational time (in s) required to fit the dimensionality model - numeric scalar

• model: Fitted model constructed via the function

Examples

#See tests/testthat directory on https://github.com/jdestefani/ExtendedDFML

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