profile.brglm: Calculate profiles for objects of class 'brglm'.

In brglm: Bias Reduction in Binomial-Response Generalized Linear Models

Description

Creates "profile.brglm" objects to be used for the calculation of confidence intervals and for plotting.

Usage

## S3 method for class 'brglm'
profile(fitted, gridsize = 10, stdn = 5,
        stepsize = 0.5, level = 0.95,
        which = 1:length(coef(fitted)), verbose = TRUE,
        zero.bound = 1e-08, scale = FALSE, ...)

Arguments

fitted	an object of class "brglm".
gridsize	as in profileModel.
stdn	as in profileModel.
stepsize	as in profileModel.
level	qchisq(level,1) indicates the range that the profiles must cover.
which	as in profileModel.
verbose	as in profileModel.
zero.bound	as in profileModel.
scale	as in profileModel.
...	further arguments passed to or from other methods.

Details

profile.brglm calculates the profiles of the appropriate objectives to be used for the construction of confidence intervals for the bias-reduced estimates (see confint.brglm for the objectives that are profiled).

Value

An object of class "profile.glm" with attribute “level” corresponding to the argument level. The object supports the methods print, plot, pairs and confint and it is a list of the components:

profilesML	a "profileModel" object containing the profiles of the ordinary deviance for the maximum likelihood fit corresponding to fitted.
profilesBR	NULL if method = "glm.fit" in brglm. If method = "brglm.fit" and pl = TRUE, profilesBR is a "profileModel" object containing the profiles of the penalized deviance for the parameters of fitted. If method = "brglm.fit" and pl = FALSE profilesBR is a "profileModel" object containing the profiles of the modified score statistic (see profileObjectives) for the parameters of fitted.

Note

Objects of class "profile.brglm" support the methods:

• printwhich prints the "level" attribute of the object, as well as the supported methods.

• confintsee confint.brglm.

• plotsee plot.profile.brglm.

• pairssee plot.profile.brglm.

Author(s)

Ioannis Kosmidis, [e-mail]

See Also

profileModel, profile.brglm.

Examples

# see example in 'confint.brglm'.

Example output

Loading required package: profileModel
'brglm' will gradually be superseded by 'brglm2' (https://cran.r-project.org/package=brglm2), which provides utilities for mean and median bias reduction for all GLMs and methods for the detection of infinite estimates in binomial-response models.

brglm documentation built on May 2, 2019, 5:20 p.m.