bhmControl | R Documentation |
Auxiliary function for bhm
fitting.
Typically only used internally by 'bhmFit', but may be used to construct a control argument to either function.
bhmControl(method = 'Bayes', interaction, biomarker.main, alpha, B, R, thin, epsilon, c.n, beta0, sigma0)
method |
choose either ‘Bayes’ for Bayes method with MCMC or ‘profile’ for profile likelihood method with Bootstrap. The default value is 'Bayes' |
interaction |
an option of fitting model with interaction term When interaction = TRUE, a predictive biomarker model will be fitted When interaction = FALSE, a prognostic biomarker model will be fitted The default value is interaction = TRUE. |
biomarker.main |
include biomarker main effect, default is TRUE |
B |
number of burn in |
R |
number of replications for Bayes meothd or number of Bootstrap for profile likelihood method |
thin |
thinning parameter for Gibbs samples, default is 2 |
epsilon |
biomarker (transformed) step length for profile likelihood method, default is 0.01 |
alpha |
significance level (e.g. alpha=0.05) |
c.n |
number of threshold (i.e. the cut point), default is 1 |
beta0 |
initial value for mean of the prior distribution of beta, default is 0 |
sigma0 |
initial value for variance of the prior distribution of beta, default is 10000 |
Control is used in model fitting of "bhm".
This function checks the internal consisitency and returns a list of value as inputed to control model fit of bhm.
Based on code from Tian Fang.
Bingshu E. Chen
bhm
## To fit a prognostic model for biomarker with two cut-points, ## 500 burn-in samples and 10000 Gibbs samples, ctl = bhmControl(interaction = FALSE, B = 500, R = 10000, c.n = 2) ## ## then fit the following model ## # fit = bhmFit(x, y, family = 'surv', control = ctl) ##
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.