controlDNPROCreg {npROCRegression}

Function used to set several parameters controlling the ROC regression fitting process

Description

Function used to set several parameters controlling the ROC regression fitting process

Usage

controlDNPROCreg(step.p = 0.02, card.P = 50, link = c("probit", "logit","cloglog"), 
kbin = 30, p = 1, seed = NULL, nboot = 500, level = 0.95, 
resample.m = c("coutcome", "ncoutcome"))

Arguments

step.p

a numeric value, defaulting to 0.02. ROC curves are calculated at a regular sequence of false positive fractions with step.p increment.

card.P

an integer value specifying the cardinality of the set of false positive fractions used in the estimation processs. By default 50.

link

a character string specifying the link function (“probit”, “logit” or “cloglog”). By default the link is the probit function.

kbin

an integer value specifying the number of binning knots. By default 30.

p

an integer value specifying the order of the local polinomial kernel estimator. By default 1.

seed

an integer value specifying the seed for the bootstrap resamples. If NULL it is initialized randomly.

nboot

an integer value specifying the number of bootstrap resamples for the construction of the confidence intervals. By default 500.

level

a real value specifying the confidence level for the confidence intervals. By default 0.95

resample.m

a character string specifying if bootstrap resampling (for the confidence intervals) should be done with or without regard to the disease status (“coutcome” or “noutcome”). In both cases, a naive bootstrap is used. By default, the resampling is done conditionally on the disease status.

Author(s)

Maria Xose Rodriguez - Alvarez and Javier Roca-Pardinas

See Also

See Also DNPROCreg

Examples

data(endosim)
# Fit a model including the interaction between age and gender.
m0 <- DNPROCreg(marker = "bmi", formula.h = "~ gender + s(age) + s(age, by = gender)", 
				formula.ROC = "~ gender + s(age) + s(age, by = gender)", 
				group = "idf_status", 
				tag.healthy = 0, 
				data = endosim, 
				control = controlDNPROCreg(card.P=50, kbin=30, step.p=0.02))
summary(m0)				
plot(m0)