energy_hica: Energy criterion

In fastHICA: Hierarchical Independent Component Analysis: a Multi-Scale Sparse Non-Orthogonal Data-Driven Basis

Description

This function implements the energy criterion defined in Secchi, Vantini, and Zanini (2013).

Usage

energy_hica(HICA.obj, maxcomp = 1, nlevel = 1, plot = FALSE)

Arguments

HICA.obj	An object provided by the function basis_hica.
maxcomp	The maximum space dimension considered.
nlevel	The number of levels analyzed. Specifically the levels from p-nlevel to p-1 are analyzed, where p is the number of variables.
plot	A logical value. If TRUE the energy is plotted.

Details

This function computes the energy according the criterion presented in Secchi, Vantini and Zanini (2013). It is useful to find the best representation. It receives in input the output of the basis_hica function.

Value

energy	A matrix with maxcomp rows and p-1 columns, where p is the number of variables. In position (i,j) it contains the energy of the best i-dimensional space for the jth level of the tree. Only the last nlevel columns are filled.
components	A matrix with maxcomp rows and p-1 columns, where p is the number of variables. In position (i,j), it contains the index of th ith basis element for jth level of the tree. Only the last nlevel columns are filled.
HICA.obj	The same object, output of the function basis_hica, provided in input.

Author(s)

Piercesare Secchi, Simone Vantini, and Paolo Zanini

References

P. Secchi, S. Vantini, and P. Zanini (2014). Hierarchical Independent Component Analysis: a multi-resolution non-orthogonal data-driven basis. MOX-report 01/2014, Politecnico di Milano.

See Also

basis_hica, similarity_hica, extract_hica

fastHICA documentation built on May 2, 2019, 9:25 a.m.