Hello all. I am trying to run PAN, multilevel
multiple imputation program, in R to impute missing
data in a longitudinal dataset. I could successfully
run the multiple imputation when I only imputed one
variable. However, when I tried to impute a
time-varying covariate as well as a response variable,
I received an error message, ?Error: subscript out of
bounds.? Can anyone tell if my commands contain any
mistakes?
First I imported SAS dataset ?sim? which includes a
response variable ?MIY1?, a time-varying covariate
?TCOV1?, TIME, GROUP (0 or 1), and ID. 200
participants were included and measurement occurred
six times. Approximately 25% of participants dropped
out at end.
sim <- read.xport('c:\\xptds.dat')
int <- rep(1,1200)
y <- cbind(sim$MIY1,sim$TCOV1)
subj <- sim$ID
pred <- cbind(int, sim$TIME, sim$GROUP)
xcol <- 1:3
zcol <- 1
prior <- list(a=2,Binv=4,c=2,Dinv=4)
pan(y,subj,pred,xcol,zcol,prior,seed=13579,iter=1000)
Error: subscript out of bounds
By the way, I also received the same error message
when I tried to include intercept and time in Zcol, a
matrix for random effect specification. I used
command ? zcol <- 1:2?. Does anybody know if this
error is due to sample size/proportion of missing data
or due to command mistake?
I truly appreciate any feedbacks.
Best regards,
Eishi