Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!
Get Started for Free

crch.control

Auxiliary Function for Controlling crch Fitting


Description

Auxiliary function for crch fitting. Specifies a list of values passed to optim.

Usage

crch.control(method = "BFGS", maxit = NULL, hessian = NULL,
  trace = FALSE, start = NULL, dot = "separate",
  lower = -Inf, upper = Inf, ...)

Arguments

method

optimization method passed to optim

maxit

the maximum number of iterations. Default is 5000 except for method="boosting" where the default is 100.

hessian

logical or NULL. If TRUE the numerical Hessian matrix from the optim output is used for estimation of the covariance matrix. If FALSE no covariance matrix is computed. If NULL (the default) the Hessian matrix is computed analytically for dist="gaussian", dist="logistic", and dist="student" with predefined df. For dist="student" without prespecified df, no analytical solution is available and a numerical Hessian matrix is forced.

trace

non-negative integer. If positive, tracing information on the progress of the optimization is produced.

start

initial values for the parameters to be optimized over.

dot

character specifying how to process formula parts with a dot (.) on the right-hand side. This can either be "separate" so that each formula part is expanded separately or "sequential" so that the parts are expanded sequentially conditional on all prior parts. Default is "separate"

lower, upper

bounds on the variables for the "L-BFGS-B" method, or bounds in which to search for method "Brent".

...

additional parameters passed to optim.

Value

A list with components named as the arguments.

See Also


crch

Censored Regression with Conditional Heteroscedasticity

v1.0-4
GPL-2 | GPL-3
Authors
Jakob Messner [aut, cre] (<https://orcid.org/0000-0002-1027-3673>), Achim Zeileis [aut] (<https://orcid.org/0000-0003-0918-3766>), Reto Stauffer [aut] (<https://orcid.org/0000-0002-3798-5507>)
Initial release
2019-08-19

We don't support your browser anymore

Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.