I have a situation in which I have an initial dataset to which subsequent datasets will be added over time. These future datasets will need to be imputed.
Is it possible with the mice package in R to add additional rows to an imputed dataset, and impute those additional rows with the same assumed distributions as the first dataset was imputed? In this case I would not want the new data to skew the distributions from which imputed values were drawn?
data <- airquality
data[4:10, 3] <- rep(NA, 7)
data[1:5, 4] <- NA
data.1 <- data[1:140, ]
data.2 <- data[141:153, ]
library(mice)
tempData <- mice(data.1, m=5, maxit=50, meth='pmm', seed=500)
summary(tempData)
# Now impute data.2 drawing from data.1's distributions.