Dear all,
I would like to understand the effect of censoring in a linear model therefore I
am doing the following
n <- 20
cen <- sample(x=c(0,1), size=n-1, replace=TRUE, prob=c(0.8,0.2))
cen <- c(cen,0)
x <- rnorm(n,0,1)
tt <- -1+x+log(-log(1-runif(n)))+0.5767
y <- c()
for(i in 1:n){
y[i] <- min(tt[i],cen[i])
}
moda <- lm(tt~x)
modb <- lm(y~x)
Is this simulation correct? Thank you in advance!
Samuel
[[alternative HTML version deleted]]