crisk.pre.bart: Data construction for competing risks with BART

Description

Competing risks contained in \((t, \delta, x)\) must be translated to data suitable for the BART competing risks model; see crisk.bart for more details.

Usage

crisk.pre.bart( times, delta, x.train=NULL, x.test=NULL,
                x.train2=x.train, x.test2=x.test, K=NULL )

Arguments

times

The time of event or right-censoring.

delta

The event indicator: 1 is a cause 1 event, 2 a cause 2 while 0 is censored.

x.train

Explanatory variables for training (in sample) data of cause 1. If provided, must be a matrix with (as usual) rows corresponding to observations and columns to variables.

x.test

Explanatory variables for test (out of sample) data of cause 1. If provided, must be a matrix and have the same structure as x.train.

x.train2

Explanatory variables for training (in sample) data of cause 2. If provided, must be a matrix with (as usual) rows corresponding to observations and columns to variables.

x.test2

Explanatory variables for test (out of sample) data of cause 2. If provided, must be a matrix and have the same structure as x.train.

If provided, then coarsen times per the quantiles \(1/K, 2/K, ..., K/K\).

Value

surv.pre.bart returns a list. Besides the items listed below, the list has a times component giving the unique times and K which is the number of unique times.

y.train

A vector of binary responses for cause 1.

y.train2

A vector of binary responses for cause 2.

cond

A vector of indices of y.train indicating censored subjects.

binaryOffset

The binary offset for y.train.

binaryOffset2

The binary offset for y.train2.

tx.train

A matrix with rows consisting of time and the covariates of the training data for cause 1.

tx.train2

A matrix with rows consisting of time and the covariates of the training data for cause 2.

tx.test

A matrix with rows consisting of time and the covariates of the test data, if any, for cause 1.

tx.test2

A matrix with rows consisting of time and the covariates of the test data, if any, for cause 2.

References

Sparapani, R., Logan, B., McCulloch, R., and Laud, P. (2016) Nonparametric survival analysis using Bayesian Additive Regression Trees (BART). Statistics in Medicine, 16:2741-53 <doi:10.1002/sim.6893>.

Examples

Run this code

# NOT RUN {
data(transplant)

delta <- (as.numeric(transplant$event)-1)

delta[delta==1] <- 4
delta[delta==2] <- 1
delta[delta>1] <- 2
table(delta, transplant$event)

table(1+floor(transplant$futime/30.5)) ## months
times <- 1+floor(transplant$futime/30.5)

typeO <- 1*(transplant$abo=='O')
typeA <- 1*(transplant$abo=='A')
typeB <- 1*(transplant$abo=='B')
typeAB <- 1*(transplant$abo=='AB')
table(typeA, typeO)

x.train <- cbind(typeO, typeA, typeB, typeAB)

N <- nrow(x.train)

x.test <- x.train

x.test[1:N, 1:4] <- matrix(c(1, 0, 0, 0), nrow=N, ncol=4, byrow=TRUE)

pre <- crisk.pre.bart(x.train=x.train, times=times, delta=delta, x.test=x.test)

# }