Control parameters for logistf

Share:

Description

Sets parameters for Newton-Raphson iteration in Firth's penalized-likelihood logistic regression

Usage

1
2
logistf.control(maxit = 25, maxhs = 5, maxstep = 5, lconv = 1e-05, gconv = 1e-05, 
    xconv = 1e-05, collapse=TRUE)

Arguments

maxit

the maximum number of iterations

maxhs

the maximum number of step-halvings in one iteration. The increment of the beta vector within one iteration is divided by 2 if the new beta leads to a decrease in log likelihood.

maxstep

specifies the maximum step size in the beta vector within one iteration.

lconv

specifies the convergence criterion for the log likelihood.

gconv

specifies the convergence criterion for the first derivative of the log likelihood (the score vector).

xconv

specifies the convergence criterion for the parameter estimates.

collapse

if TRUE, evaluates all unique combinations of x and y and collapses data set. This may save computing time with large data sets with only categorical (binary) covariates.

Details

logistf.control() is used by logistf and logistftest to set control parameters to default values. Different values can be specified, e. g., by logistf(..., control= logistf.control(maxstep=1)).

Value

maxit

the maximum number of iterations

maxhs

the maximum number of step-halvings in one iteration. The increment of the beta vector within one iteration is divided by 2 if the new beta leads to a decrease in log likelihood.

maxstep

specifies the maximum step size in the beta vector within one iteration.

lconv

specifies the convergence criterion for the log likelihood.

gconv

specifies the convergence criterion for the first derivative of the log likelihood (the score vector).

xconv

specifies the convergence criterion for the parameter estimates.

collapse

if TRUE, evaluates all unique combinations of x and y and collapses data set.

Author(s)

Georg Heinze

Examples

1
2
3
4
data(sexagg)
fit2<-logistf(case ~ age+oc+vic+vicl+vis+dia, data=sexagg, weights=COUNT, 
   control=logistf.control(maxstep=1))
summary(fit2)