So I am dealing with the imputation of data set having time to event data. Several papers suggest that the use of Nelson-Aalen estimate (as an approximation to the baseline hazard) will provide better imputation results. Is there a way to find Nelson-Aalen estimate and bind to my data set in R. I have found a function named nelsonaalen(data, time, event) in the mice package but I am afraid whether it will cause any error since I can only include one time variable(failure time) in it. The variables in my data are as follows:
N = 1000
xt=runif(N, 0, 50)
x1=rnorm(N, 2, 1)
x2=rnorm(N, -2, 1)
x3 <- rnorm(N, 0.5*x1 + 0.5*x2, 2)
x4 <- rnorm(N, 0.3333*x1 + 0.3333*x2 + 0.3333*x3, 2 )
lp <- 0.05*x1 + 0.2*x2 + 0.1*x3 + 0.02*x4
T <- qweibull(runif(N,pweibull(xt,shape = 7.5, scale = 84*exp(-lp/7.5)),1), shape=7.5, scale=84*exp(-lp/7.5))
Cens1 <- 100
time_M <- pmin(T,Cens1)
event_M <- time_M == Tm
Here xt denotes starting time, T denotes the failure time and the x1 to x4 are my covariates in which I'll create missing values in two of the covariates (x3 and x4).