gldrm.control: Control arguments for 'gldrm' algorithm
In gldrm: Generalized Linear Density Ratio Models

gldrm.control

R Documentation

Control arguments for `gldrm` algorithm

Description

This function returns control arguments for the gldrm algorithm. Each argument has a default value, which will be used unless a different value is provided by the user.

Usage

gldrm.control(
  eps = 1e-10,
  maxiter = 100,
  returnfTiltMatrix = TRUE,
  returnf0ScoreInfo = FALSE,
  print = FALSE,
  betaStart = NULL,
  f0Start = NULL
)

Arguments

`eps`	Convergence threshold. The fitting algorithm has converged when the relative change in log-likelihood between iterations is less than `eps`. A single iteration consists of a `beta` update followed by an `f0` update.
`maxiter`	Maximum number of iterations allowed.
`returnfTiltMatrix`	Logical. Return nonparametric fitted probabilities for each observation. This is a matrix with nrow equal to the number of observations and ncol equal to the number of unique response values observed.
`returnf0ScoreInfo`	Logical. If `TRUE`, the score and information for `log(f0)` are returned as components of the "gldrm" object.
`print`	Logical. If `TRUE`, the relative change in the log-likelihood will be printed after each iteration.
`betaStart`	Optional vector of starting values for `beta`. If the call to gldrm contains a formula, the values of betaStart should correspond to the columns of the model matrix.
`f0Start`	Optional vector of starting values for `f0`. The length of the vector should be the number of unique values in the response, and the vector should correspond to these values sorted in increasing order. The starting values will be scaled to sum to one and tilted to have mean `mu0`. All values should be strictly positive.