R/noise.R

In AntoineDubois/RUMdesignSimulator: RUM design Simulator

#############################################################
## This file enlists functions which generate noise
## characteristics.
## The notations are taken from Train 2003.
## antoine.dubois.fr@gmail.com - March 2021
#############################################################

##############################
# 1 - Decision makers' observed characteristics generation
##############################
#' @title ob_decision_makers_att
#'
#' @description The class of observed decision makers' attributes
#'
#' @param N The number of decision makers
#'
#' @param p The number of attributes of each decision makers
#'
#' @method gen generates the attributes by the chosen distribution
#'
#' @examples N <- 10; p <- 3
#' a <- ob_decision_makers_att$new(N=N, p=p);
#' a$gen("student", param=list(location=rep(100, 3), df=3))
#' a$gen()
#' a$gen("help")
#'
#' @export factorial design
#'
#' @title noise
#'
#' @description The class noise
#'
#' @param N The number of decision makers
#'
#' @param J The number of attributes of alternatives
#'
#' @method gen generates the noise
#'
#'
#' @examples N <- 10; J <- 5; a <- noise$new(N=N, J=J);
#'           a$gen()
#'           a$gen("normal", param=list(mu=rep(-1000, J)))
#'           a$gen("student", param=list(scale=diag(2, J),df=10))
#'           a$gen("discrete_uniform")
#'
#' @export factorial design
#'

noise <- setRefClass("noise", fields = list(N="numeric", J="numeric"),
                                      methods = list(gen=function(law="gumbel", n=N, m=J, param=list()){generation(law, n, m, param)}))