geometric_mean: Computation of SCMs' geometric mean

In LauraMasiero/RiemannR: RiemannR - EEG-based BCI analysis with Riemannian geometry

Description

This function allows to compute the geometric mean of the given sample covariance matrices.

Usage

geometric_mean(scm_list, eps=1/3)

Arguments

scm_list	a list of numeric sample covariance matrices.
eps	parameter in [0,1] that controls the convergence speed. Higher eps values lead to a faster convergence, but the more close to 1 it is, the more irrealistic the underlining hypothesis are (eps=1 indicates data without ANY noise).

Value

It returns a list containing the geometric mean.

Author(s)

Laura Masiero, email: [e-mail]

References

The algorithm used to compute the geometric mean can be found here: Moakher, Maher (2005). "A differential geometric approach to the geometric mean of symmetric positive-definite matrices". In: SIAM Journal on Ma- trix Analysis and Applications 26.3, pp. 735–747. issn: 08954798. doi: 10.1137/S0895479803436937.

LauraMasiero/RiemannR documentation built on Sept. 29, 2020, 9:51 p.m.