replace_overimputes

Replace all overimputed data points in the mice imputation
of one variable. Overimputed data points are those data
that were not missing in the original but were marked for
imputation manually and imputed by the imputation procedure.

Creates a uniform interface for several advanced imputations missing data methods. Every available method can be used as a part of 'mlr3' pipelines which allows easy tuning and performance evaluation. Most of the used functions work separately on the training and test sets (imputation is trained on the training set and impute training data. After that imputation is again trained on the test set and impute test data).

Jan Borowski

NADIA

NA Data Imputation Algorithms

replace_overimputes function

<dl><dt>data</dt>
<dd>data.frame the original, non-imputed dataset (mids$data)</dd>
<dt>imp</dt>
<dd>list of data.frames all imputations stored in the mids object</dd>
<dt>j</dt>
<dd>character scalar the name of the variable whose imputations should be replaced</dd>
<dt>i</dt>
<dd>character or integer scalar the number of the current imputation (can be 1:m)</dd></dl>

Arguments

Replace overimputes. Used in mice.reuse. — replace_overimputes

<dl>

<dt>data</dt>
<dd>data.frame the original, non-imputed dataset (mids$data)</dd>


<dt>imp</dt>
<dd>list of data.frames all imputations stored in the mids object</dd>


<dt>j</dt>
<dd>character scalar the name of the variable whose imputations should be replaced</dd>


<dt>i</dt>
<dd>character or integer scalar the number of the current imputation (can be 1:m)</dd>

</dl>

replace_overimputes: Replace overimputes. Used in mice.reuse.

Description

Usage

Arguments