Hi, I'm trying to impute a data set consisting of mixed type variables, mostly logical, but also ordered and non-ordered factors, and numeric variables with the missForest package. Although the help file of missForest states that the resulting data matrix 'ximp' has the same type as the original data 'xmis', the ximp I get has all numeric variables with decimal numbers. Has someone any explanations for this behavior? Can the data types of ximp be controlled? If not, is there an easy way to coerce the variables back to their original data types by rounding to the nearest levels of the factors? Thanks Nima -- View this message in context: http://r.789695.n4.nabble.com/Package-missForest-changes-data-types-tp3766921p3766921.html Sent from the R help mailing list archive at Nabble.com.