TML1.noncensored.control.S: Control parameters for S-estimate of location and scale

In RobustAFT: Truncated Maximum Likelihood Fit and Robust Accelerated Failure Time Regression for Gaussian and Log-Weibull Case

Control parameters for S-estimate of location and scale

Description

Auxiliary function for TML1.noncensored. Typically only used internally by TML1.noncensored, but may be used to construct a control argument. This function provides default values.

Usage

TML1.noncensored.control.S(tlo = 1e-04, mxf = 50, mxs = 50, ntm = 50,
                           tls = 1e-06, h = 100)

Arguments

tlo	Relative tolerance in the iterative algorithms.
mxf	Maximum number of iterations in computing the location estimate.
mxs	Maximum number of iterations in computing the scale estimate.
ntm	Parameter used in iteration monitoring. When the number of iterations is a multiple of ntm, the current parameter values are printed.
tls	Tolerance for denominators. If a scale estimate is less than tls, the scale estimate is set equal to tls.
h	The number of subdivisions of the interval (min(yi), max(yi)) used is the computation of the estimate \lambda^(0).

Value

List containing the desired values for each of the control parameters, plus the value Beta0 of \beta.

RobustAFT documentation built on Aug. 21, 2023, 5:13 p.m.