[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
I did not want to say, that Morton should not contact Patrick Royston
if there is a problem in -ice-'s -conditional- option.
I did not want to imply that you will never need to specify conditions.
I understood from the question, that imputations are to be done
conditional on variables, which are not themselevs imputed. For this
purpose I think the distiction between hard and soft missing values
(which Stata's -mi- offers, while -ice- does not) is sufficient and
you do not need conditional imputations here.
* For searches and help try: