RCdata: Simulated Right Censored Data Set for a 5 State Model
In msSurv: Nonparametric Estimation for Multistate Models

Description Usage Format Details References Examples

Simulated data from a five-state progressive model with both branching and tracking between states. Possible transitions are 1->2, 1->3, 2->4, and 2->5, with states 3, 4, and 5 as terminal states.

1	data(RCdata)

A data frame with 1000 individuals with the following 4 variables.

id: Identification number
stop: Transition time
start.stage: State transitioning FROM
end.stage: State transitioning TO

A data set of 1000 individuals subject to independent right censoring was simulated, with 60% of individuals starting in state 1 at time 0 and 40% starting in state 2. Those in state 1 remained there until they transitioned to the transient state 2 or the terminal state 3. Individuals in state 2 remained there until they transitioned to either terminal state 4 or 5.

Nicole Ferguson, Somnath Datta, Guy Brock (2012). msSurv: An R Package for Nonparametric Estimation of Multistate Models. Journal of Statistical Software, 50(14), 1-24. URL http://www.jstatsoft.org/v50/i14/.

data(RCdata)

####################################################################
##               Code used to generate data                       ##
####################################################################

## Specifying the tree structure for the simulation

Nodes <- c("1", "2", "3", "4", "5") ## states possible in MSM
Edges <- list("1"=list(edges=c("2", "3")), "2"=list(edges=c("4", "5")),
           "3"=list(edges=NULL), "4"=list(edges=NULL), "5"=list(edges=NULL))
RCtree <- new("graphNEL", nodes=Nodes, edgeL=Edges, edgemode="directed")

## Simulating the data
set.seed(123)
n <- 1000
n1 <- 0.6*n
n2<-0.4*n
ill <- round(rweibull(n1, 2), digits=4)
death1 <- round(rweibull(n1, 2), digits=4)
allcensor <- round(rlnorm(n, meanlog=-0.5, sdlog=2), digits=4)
censor<-allcensor[1:n1]
censor2<-allcensor[(1:n2)+length(n1)]

c1 <- pmin(ill, death1, censor)
num.ill <- sum(ill<death1  & ill<censor) ## number who are "ill"
stage <- ifelse(censor<death1 & censor<ill, 0, ifelse(ill<death1, 2, 3))
data1 <- data.frame(id=1:length(c1), stop=c1, start.stage=1, end.stage=stage)

ind <- which(data1$end.stage==2) ## those transitioning to state 2
## no.st2 <- sum(as.numeric(data1$end.stage==2)) ## number that made transition to 2
death24 <- round(qweibull(pweibull(ill, shape=2) + runif(n1, 0, 1)*
                              (1-pweibull(ill, shape=2)), shape=2), digits=4)

death25 <- round(qweibull(pweibull(ill, shape=2) + runif(n1, 0, 1)*
                              (1-pweibull(ill, shape=2)), shape=2), digits=4)
c2 <- pmin(death24[ind], death25[ind], censor[ind])
stage[ind] <- ifelse(censor[ind]<death24[ind] & censor[ind]<death25[ind], 0,
                     ifelse(death24[ind]<death25[ind], 4, 5))
data2 <- data.frame(id=ind, stop=c2, start.stage=data1$end.stage[ind],
                    end.stage=stage[ind])

death24.2 <- round(rweibull(n2, 2), digits=4)
death25.2 <- round(rweibull(n2, 2), digits=4)
tran<-pmin(death24.2, death25.2, censor2)
stage2 <- ifelse(censor2<death24.2 & censor2<death25.2, 0,
                 ifelse(death24.2<death25.2, 4, 5))
data2.2 <- data.frame(id=(1:n2)+length(c1), stop=tran, start.stage=2, end.stage=stage2)

RCdata <- rbind(data1, data2, data2.2)
RCdata <- RCdata[order(RCdata$id, RCdata$stop), ]