data {bhm}

dataset

Description

dataset for biomarker threshold model (bhm)

Usage

# to generate survival data, use: 

   gendat.surv(n, c0, beta, type=c("brm", "bhm"))

# to generate glm data, use:

   gendat.glm(n, c0, beta)

Arguments

n

sample size

c0

cut off point, for example c0 = 0.4

beta

regression coefficient, for example, beta = c(0.3, log(0.5), log(0.25))

type

type of biomarker threshold model, either bhm or brm, default is type = "brm"

Format

The format of the data set for analysis shall be a data frame with a response variable (either a Surv object for Cox model or a glm response variable object) and at least one dependent variable as the biomarker variable.

Details

data set of prostate cancer in the 'survival' package is used as an example in paper by Chen, et al. (2014).

Source

prosate dataset can be loaded with 'library(survival)'.

References

Chen, B. E., Jiang, W. and Tu, D. (2014). A hierarchical Bayes model for biomarker subset eff ects in clinical trials. Computational Statistics and Data Analysis. vol 71, page 324-334.

Examples

#data(data)
## maybe str(data) ; plot(data) ...
c0 = 0.4
b = c(-0.5, 1.5, 1.3)
data = gendat.surv(n=30, c0 = c0, beta = b)