7
$\begingroup$

What is Python's alternative to missing data imputation with mice in R? Imputation using median/mean seems pretty lame, I'm looking for other methods of imputation, something like randomForest.

$\endgroup$
  • $\begingroup$ You might want to take a look at fancyimpute $\endgroup$ – chainD Jun 30 '17 at 3:19
1
$\begingroup$

You can use the library HERE. You are able to impute the values of your dataset using linear models with an Sklearn type interface.

| improve this answer | |
$\endgroup$
  • 1
    $\begingroup$ Instead of requesting contributions to your repository, you could edit the answer and include more details about the current features of the library. $\endgroup$ – tuomastik Jul 26 '17 at 3:51
1
$\begingroup$

Is this statsmodels MICE implementation an option? I also heard that an implementation of MICE is being merged into the development branch of scikit-learn but I can't find any info on that.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.