Validating a nomogram for a survey-weighted Cox model

Validates a nomogram for a survey-weighted Cox model using bootstrap.

Usage

svycox.validate(.boot.index, .nom, .data)

Arguments

.boot.index: a matrix of bootstrap sample indicators with the number of rows the same as the number of rows in the data on which the nomogram was created and the number of columns being the number of bootstrap samples
.nom: a nomogram object returned from svycox.nomogram
.data: contains the dataset on which the validation will take place

Details

Note that generating the bootstrap sample is design dependent and it is not part of the function. The user has to generate the bootstrap samples consistent with the design used. An example of how the bootstrap sample was generated for the dataset is presented in the reference below.

Value

prints the estimated optimism and returns the vector of optimism values for each bootstrap sample which can be used to summarize the validation with the measure of choice

References

Marinela Capanu, Mithat Gonen (2015). Building a Nomogram for Survey-Weighted Cox Models Using R. Journal of Statistical Software, Code Snippets, 64(1), 1-17. URL https://www.jstatsoft.org/v64/c01/.

Examples

bootit=200
library(survey)
library(rms)
data(noNA)
dd=datadist(noNA)
options(datadist="dd")
dstr2=svydesign(id=~1, strata=~group, prob=~inv_weight, fpc=~ssize, data=noNA)
mynom=svycox.nomogram(.design=dstr2, .model=Surv(survival,surv_cens)~ECOG+liver_only+Alb+Hb+Age+
Differentiation+Gt_1_m1site+lymph_only, .data=noNA, pred.at=24, fun.lab="Prob of 2 Yr OS")
#> Error in Design(X, formula = formula): dataset dd not found for options(datadist=)

cases=which(noNA$group=="long")
controls=which(noNA$group=="<24")
boot.index=matrix(NA,nrow(noNA),bootit)
for(i in 1:bootit){
boot.index[,i]=c(sample(cases,replace=TRUE),sample(controls,replace=TRUE))
}
myval=svycox.validate(boot.index,mynom,noNA)  
#> Error in paste("~", names(.nom2$design$prob)): object 'mynom' not found