mc_build_C: Build the joint covariance matrix
In mcglm: Multivariate Covariance Generalized Linear Models

mc_build_C

R Documentation

Build the joint covariance matrix

Description

This function builds the joint variance-covariance matrix using the Generalized Kronecker product and its derivatives with respect to rho, power and tau parameters.

Usage

mc_build_C(
  list_mu,
  list_Ntrial,
  rho,
  list_tau,
  list_power,
  list_Z,
  list_sparse,
  list_variance,
  list_covariance,
  list_power_fixed,
  compute_C = FALSE,
  compute_derivative_beta = FALSE,
  compute_derivative_cov = TRUE
)

Arguments

`list_mu`	A list with values of the mean.
`list_Ntrial`	A list with the number of trials. Usefull only for binomial responses.
`rho`	Vector of correlation parameters.
`list_tau`	A list with values for the tau parameters.
`list_power`	A list with values for the power parameters.
`list_Z`	A list of matrix to be used in the matrix linear predictor.
`list_sparse`	A list with Logical.
`list_variance`	A list specifying the variance function to be used for each response variable.
`list_covariance`	A list specifying the covariance function to be used for each response variable.
`list_power_fixed`	A list of Logical specifying if the power parameters are fixed or not.
`compute_C`	Logical. Compute or not the C matrix.
`compute_derivative_beta`	Logical. Compute or not the derivative of C with respect to regression parameters.
`compute_derivative_cov`	Logical. Compute or not the derivative of C with respect the covariance parameters.

Value

A list of matrices. This function return the variance-covariance matrix C or its inverse along with its respective matrices of derivatives with respect to rho, power and tau parameters. The returned object is intended for internal use only.