For each value in the prediction vector <code>xtest</code>, one of the closest <code>k</code>
values in the prediction vector <code>xtrain</code> is randomly chosen and its observed
value in <code>ytrain</code> is returned. Note that <code>xtrain</code> and <code>xtest</code> must be both either
numeric, logical, or factor-valued. <code>ytest</code> can be of any type.

Alternative implementation of the beautiful 'MissForest'
algorithm used to impute mixed-type data sets by chaining random
forests, introduced by Stekhoven, D.J. and Buehlmann, P. (2012)
<doi:10.1093/bioinformatics/btr597>. Under the hood, it uses the
lightning fast random forest package 'ranger'. Between the iterative
model fitting, we offer the option of using predictive mean matching.
This firstly avoids imputation with values not already present in the
original data (like a value 0.3334 in 0-1 coded variable). Secondly,
predictive mean matching tries to raise the variance in the resulting
conditional distributions to a realistic level. This would allow,
e.g., to do multiple imputation when repeating the call to
missRanger(). Out-of-sample application is supported as well.

Michael Mayer

missRanger

Fast Imputation of Missing Values

pmm function

<dl><dt>xtrain</dt>
<dd>Vector with predicted values in the training data.
Must be numeric, logical, or factor-valued.</dd>
<dt>xtest</dt>
<dd>Vector as <code>xtrain</code> with predicted values in the test data.
Missing values are not allowed.</dd>
<dt>ytrain</dt>
<dd>Vector of the observed values in the training data. Must be of same
length as <code>xtrain</code>.</dd>
<dt>k</dt>
<dd>Number of nearest neighbours (donors) to sample from.</dd>
<dt>seed</dt>
<dd>Integer random seed.</dd></dl>

Arguments

Predictive Mean Matching — pmm

<dl>

<dt>xtrain</dt>
<dd>Vector with predicted values in the training data.
Must be numeric, logical, or factor-valued.</dd>


<dt>xtest</dt>
<dd>Vector as <code>xtrain</code> with predicted values in the test data.
Missing values are not allowed.</dd>


<dt>ytrain</dt>
<dd>Vector of the observed values in the training data. Must be of same
length as <code>xtrain</code>.</dd>


<dt>k</dt>
<dd>Number of nearest neighbours (donors) to sample from.</dd>


<dt>seed</dt>
<dd>Integer random seed.</dd>

</dl>

pmm: Predictive Mean Matching

Description

Usage

Value

Arguments

Examples