SemiCompRisks (version 2.4)

simID: The function that simulates independent/cluster-correlated semi-competing risks data under semi-Markov Weibull/Weibull-MVN models.

Description

The function to simulate independent/cluster-correlated semi-competing risks data under semi-Markov Weibull/Weibull-MVN models.

Usage

simID(cluster=NULL, x1, x2, x3, beta1.true, beta2.true, beta3.true, 
			alpha1.true, alpha2.true, alpha3.true, 
			kappa1.true, kappa2.true, kappa3.true, 
			theta.true, SigmaV.true=NULL, cens)

Arguments

cluster

a vector of cluster information for n subjects. The cluster membership must be set to consecutive positive integers, $1:J$. Required only when generating clustered data.

x1

covariate matrix, n observations by p1 variables.

x2

covariate matrix, n observations by p2 variables.

x3

covariate matrix, n observations by p3 variables.

beta1.true

true value for $\beta_1$.

beta2.true

true value for $\beta_2$.

beta3.true

true value for $\beta_3$.

alpha1.true

true value for $\alpha_1$.

alpha2.true

true value for $\alpha_2$.

alpha3.true

true value for $\alpha_3$.

kappa1.true

true value for $\kappa_1$.

kappa2.true

true value for $\kappa_2$.

kappa3.true

true value for $\kappa_3$.

theta.true

true value for $\theta$.

SigmaV.true

true value for $\Sigma_V$. Required only when generating clustered data.

cens

a vector with two numeric elements. The right censoring times are generated from Uniform($cens[1]$, $cens[2]$).

Value

• simIDcor returns a data.frame containing semi-competing risks outcomes from n subjects. It is of dimension $n\times 4$: the columns correspond to $y_1$, $\delta_1$, $y_2$, $\delta_2$.
• y1
  a vector of n times to the non-terminal event
• y2
  a vector of n times to the terminal event
• delta1
  a vector of n censoring indicators for the non-terminal event time (1=event occurred, 0=censored)
• delta2
  a vector of n censoring indicators for the terminal event time (1=event occurred, 0=censored)

Examples

library(MASS)
	set.seed(123456)
	
	J = 110
	nj = 50
	n = J * nj

	cluster <- rep(1:J, each = nj)

	kappa1.true <- 0.05
	kappa2.true <- 0.01
	kappa3.true <- 0.01
	alpha1.true <- 0.8
	alpha2.true <- 1.1
	alpha3.true <- 0.9
	beta1.true <- c(0.5, 0.8, -0.5)
	beta2.true <- c(0.5, 0.8, -0.5)
	beta3.true <- c(1, 1, -1)
	SigmaV.true <- matrix(0.25,3,3)
	theta.true <- 0.5
	cens <- c(90, 90)

	cov1 <- matrix(rnorm((length(beta1.true)-1)*n, 0, 1), n, length(beta1.true)-1)
	cov2 <- sample(c(0, 1), n, replace = TRUE)
	x1 <- as.data.frame(cbind(cov1, cov2))
	x2 <- as.data.frame(cbind(cov1, cov2))
	x3 <- as.data.frame(cbind(cov1, cov2))
	
	simData <- simID(cluster, x1, x2, x3, beta1.true, beta2.true, beta3.true, 
				alpha1.true, alpha2.true, alpha3.true, 
				kappa1.true, kappa2.true, kappa3.true, 
				theta.true, SigmaV.true, cens)