nll2: Negative log-likelihood

In rje42/causl: Methods for Specifying, Simulating from and Fitting Causal Models

Negative log-likelihood

Description

Negative log-likelihood

Usage

nll2(
  theta,
  dat,
  mm,
  beta,
  phi,
  inCop,
  fam_cop = 1,
  family,
  link,
  cop_pars = NULL,
  use_cpp = TRUE,
  other_pars = list()
)

Arguments

theta	concatenated vector of parameters (beta followed by phi)
dat	matrix of data
mm	model matrix for use with beta
beta	(sparse) matrix of regression parameters for each variable and copula
phi	vector of dispersion parameters
inCop	vector of integers giving variables in dat to be included in copula
fam_cop, family	integer and integer vector for copula and distribution families respectively
link	vector of link functions
cop_pars	other parameters for copula
use_cpp	logical: should Rcpp functions be used?
other_pars	other parameters to pass to glm_dens

Details

The number of columns of beta should be the number of columns in dat plus the number required to parameterize the copula. The first few columns and the entries in phi are assumed to be in the order of those in dat. If the ith family for a variable does not require a dispersion parameter then the value of phi[i] is ignored.

rje42/causl documentation built on June 1, 2025, 2:50 p.m.