R/effeihgcov.R

In CUB: A Class of Mixture Models for Ordinal Data

#' @title Auxiliary function for the log-likelihood estimation of IHG models with covariates
#' @description Compute the opposite of the log-likelihood function for an IHG model with covariates
#' for the preference parameter.  
#' @aliases effeihgcov
#' @usage effeihgcov(nu, ordinal, U, m)
#' @param  nu Vector of the starting values for the parameters to be estimated, with length equal to  
#'  NCOL(U)+1 to account for an intercept term (first entry of \eqn{nu})
#' @param ordinal Vector of ordinal responses
#' @param U Matrix of the explanatory variables for the preference parameter \eqn{\theta}
#' @param m Number of ordinal categories
#' @keywords internal 
#' @details It is called as an argument for "optim" within IHG function (with covariates)
#'  as the function to minimize.

effeihgcov <-
function(nu,ordinal,U,m){
  -loglikihgcov(m,ordinal,U,nu)
}