I'm trying to understand MICE imputation process, I have read some articles and I have understood how the imputation happens, but I didn't get the pooling step. After analyzing the resulting datasets with Rubin's rules, how to pool these datasets? How to get only one dataset?

In the end, do I combine all these datasets? If yes, how? Or do I compare every dataset's estimators with Rubin's estimators and choose one dataset?

  • 1
    $\begingroup$ Hi and welcome to this community! What is "MICE imputation"? Where did you hear of it? $\endgroup$ – nbro Jan 8 at 12:20
  • $\begingroup$ hello, it's a statistical method for imputing missing data (Multiple Imputation with Chained Equation) it uses regression , thank you i have already found the answer, we take the last converged dataset and to combine estimators such as variance we use rubin rules $\endgroup$ – Asmaa Jan 17 at 23:07
  • $\begingroup$ You can give an answer below to your own question. Your answer may be beneficial to future readers! $\endgroup$ – nbro Jan 17 at 23:45
  • $\begingroup$ yes of course, i have already mentionned the answer in the comment. $\endgroup$ – Asmaa Jan 19 at 12:43

Your Answer

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

Browse other questions tagged or ask your own question.