MRH (version 2.2)

nph: INTERNAL FUNCTION NEEDED FOR NON-PROPORTIONAL HAZARDS MODELS.

Description

nph is used in the MRH survival model formula to denote that a variable should be modeled under the non-proportional hazards assumption.

Usage

nph(x)

Arguments

x

x is the covariate that will be modeled under the non-proportional hazards assumption. It must be a categorical variable, although it need not be entered using the "factor" class in R. A separate hazard rate will be estimated for each group in x.

References

Dukic, V., Dignam, J. (2007), Bayesian hierarchical multiresolution hazard model for the study of time-dependent failure patterns in early stage breast cancer. Bayesian Analysis. 2, 591--610.

Examples

# Examine the tongue data set (from the KMest package), and
# quantify how the rumor DNA profile (1=Aneuploid Tumor, 2=Diploid Tumor)
# affects survival time (in weeks).
data(tongue)

# Fit the MRH model, including tumor type using the 
# non-proportional hazards model.  
# With 16 bins (M = 4), each bin represents 25 weeks.
# Generally it is recommended to use a higher burn-in value, 
# thinning value, and maximum number
# of iterations, but for illustrative purposes these values 
# have been lowered.
## Not run: 
# fit.tongue = estimateMRH(formula = Surv(time, delta) ~ 
# 	nph(type), data = tongue, M = 4, 
# 	burnIn = 200, maxIter = 2000, thin = 1, 
# 	outfolder = 'MRH_tongue_nph')## End(Not run)