missForest is a nonparametric, mixed-type imputation method for basically any type of data.
Here, we host the R-package “missForest” for the statistical software R.

The method is based on the publication Stekhoven and Bühlmann, 2012. The R package contains a vignette on how to use “missForest” in R including many helpful examples.


Currently, we are working on going multiple imputation with missForest. We are testing several ways of doing it, including an implicit approach making multiple full data set imputations potentially unnecessary. We expect to see these first extensions in the second half of 2019.

Deal with tibbles and variable attributes.

Potential innovations alongside:

For later…

Stuff we consider less interesting - write us if you disagree:

Contact us

Contact me by email: stekhoven@nexus.ethz.ch

References: Stekhoven, D.J. and Buehlmann, P. (2012), ‘MissForest - nonparametric missing value imputation for mixed-type data’, Bioinformatics, 28(1) 2012, 112-118, doi: 10.1093/bioinformatics/btr597