rcm_em_step_arma: The RCM EM-step

In AEBilgrau/correlateR: Fast correlations and covariances

Description

A armadillo-based function to perform the E and M step in the EM algorithm of the RCM. This functions assumes nu to be fixed.

Usage

rcm_em_step_arma(Psi, nu, S_list, ns)

Arguments

Psi	A numeric matrix.
nu	A numeric of length 1 giving the degrees of freedom in the RCM.
S_list	A list of scatter matrices for each dataset/group of the same size a Psi.
ns	A numeric vector the same lengths as S_list giving the number of samples for each dataset.

Value

A numeric matrix the same size as Psi giving the updated Psi.

Examples

ns <-  c(5, 5, 5)
S <- createRCMData(ns = ns, psi = diag(4), nu = 30)
correlateR:::rcm_em_step_arma(Psi = diag(4), nu = 15, S_list = S, ns = ns)

AEBilgrau/correlateR documentation built on Nov. 15, 2019, 9:21 a.m.