mc_expm: Exponential-matrix covariance link function

In mcglm: Multivariate Covariance Generalized Linear Models

Exponential-matrix covariance link function

Description

Given a matrix U the function mc_expm returns the exponential-matrix expm(U) and some auxiliares matrices to compute its derivatives. This function is based on the eigen-value decomposition it means that it is very slow.

Usage

mc_expm(U, n = dim(U)[1], sparse = FALSE, inverse = FALSE)

Arguments

U	A matrix.
n	A number specifing the dimension of the matrix U. Default n = dim(U)[1].
sparse	Logical defining the class of the output matrix. If sparse = TRUE the output class will be 'dgCMatrix' if sparse = FALSE the class will be 'dgMatrix'.
inverse	Logical defining if the inverse will be computed or not.

Value

A list with Ω = expm(U) its inverse (if inverse = TRUE) and auxiliares matrices to compute the derivatives.

Author(s)

Wagner Hugo Bonat

See Also

expm, eigen, link[mcglm]{mc_dexp_gold}.

mcglm documentation built on Sept. 16, 2022, 1:06 a.m.