ug_control: Optimization Control Parameters Passed to optim

View source: R/4_maximum_likelihood_estimation.R

ug_controlR Documentation

Optimization Control Parameters Passed to optim

Description

Optimization parameters passed to optim for the maximum likelihood estimation in the unit gamma regression with parametric link functions. This function acts in the same spirit as betareg.control from the betareg package. Its primary purpose is to gather all the optimization control arguments in a single function.

Usage

ug_control(
  method = "BFGS",
  maxit = 10000,
  start = NULL,
  trace = FALSE,
  hessian = FALSE,
  ...
)

Arguments

method

the method to be used. See "Details" in optim. The default method ("BFGS") is a quasi-Newton method (also known as a variable metric algorithm), specifically that published simultaneously in 1970 by Broyden, Fletcher, Goldfarb and Shanno.

maxit

the maximum number of iterations of the algorithm. Defaults to 10000.

start

An optional vector of initial values to start the algorithm iterations. The input order should be (beta, gamma, lambda1, lambda2), where beta is a k-dimensional vector with the coefficients associated with the mean, gamma is an l-dimensional vector with the coefficients associated with the dispersion parameter, and lambda1 and lambda2 are the parameters associated with the parametric link functions, if any. Initial guesses for lambda1 or lambda2 should only be passed if the mean-related or dispersion-related link function is a parametric link function.

trace

non-negative integer. If positive, tracing information on the progress of the optimization is produced. Higher values may produce more tracing information: for method "L-BFGS-B" there are six levels of tracing. (To understand exactly what these do see the source code: higher levels give more detail.)

hessian

logical; it specifies whether the asymptotic covariance matrix of the estimates should be obtained based on the numerically differentiated Hessian provided by optim. If FALSE, Fisher's expected information matrix is used.

...

further arguments passed to optim.

Value

A list with the arguments specified.

Author(s)

Rodrigo M. R. de Medeiros <rodrigo.matheus@live.com>


rdmatheus/ugrpl documentation built on July 13, 2024, 10:24 p.m.