psglm {bcROCsurface}

Fitting verification models

Description

psglm is used to fit generalized linear models to the verification process. This function requires a symbolic formula of the linear predictor, and a specified regression model.

Usage

psglm(formula, data, model = "logit", test = FALSE, trace = TRUE, ...)

Arguments

formula

an object of class "formula": a symbolic description of the model to be fitted.

data

an optional data frame containing the variables in the model.

model

a specified model to be used in the fitting. The suggestion regression models are logit, probit and threshold. If model is ignored, then psglm use a default model as logit.

test

a logical value indicating whether p-values of the regression coefficients should be returned.

trace

switch for tracing estimation process. Default TRUE.

...

optional arguments to be passed to glm.

Details

psglm estimates the verification probabilities of the patients. The suggestion model is designed as a list containing: logit, probit and threshold.

Value

psglm returns a list containing the following components:

coeff

a vector of estimated coefficients.

values

fitted values of the model.

Hess

the Hessian of the measure of fit at the estimated coefficients.

x

a design model matrix.

formula

the formula supplied.

model

the model object used.

See Also

glm

Examples

data(EOC)
out <- psglm(V ~ CA125 + CA153 + Age, data = EOC, test = TRUE)