new_vimmi

internal

Provides methods for imputation and visualization of
missing values. It includes graphical tools to explore the amount, structure
and patterns of missing and/or imputed values, supporting exploratory
data analysis and helping to investigate potential missingness mechanisms
(details in Alfons, Templ and Filzmoser, <doi:10.1007/s11634-011-0102-y>).
The quality of imputations can be assessed visually using a wide range of
univariate, bivariate and multivariate plots.
The package further provides several imputation methods,
including efficient implementations of k-nearest neighbour and hot-deck
imputation (Kowarik and Templ 2013, <doi:10.18637/jss.v074.i07>),
iterative robust model-based multiple
imputation (Templ 2011, <doi:10.1016/j.csda.2011.04.012>;
Templ 2023, <doi:10.3390/math11122729>), and machine learning–based
approaches such as robust GAM-based multiple imputation
(Templ 2024, <doi:10.1007/s11222-024-10429-1>) as well as random forest
and gradient boosting (XGBoost) imputation
(Niederhametner et al., <doi:10.1177/18747655251339401>).
General background and practical guidance on imputation are provided in the
Springer book by
Templ (2023) <doi:10.1007/978-3-031-30073-8>.

Matthias Templ

Visualization and Imputation of Missing Values

Alexander Kowarik

Andreas Alfons

Johannes Gussenbauer

Nina Niederhametner

Eileen Vattheuer

Gregor de Cillia

Bernd Prantner

Wolfgang Rannetbauer

new_vimmi function

<dl><dt>data</dt>
<dd>Original data.frame/data.table with NAs intact</dd>
<dt>imp</dt>
<dd>Named list: per variable with missings, a data.frame with
nmis rows and m columns of imputed values</dd>
<dt>where</dt>
<dd>Logical matrix indicating which cells were imputed</dd>
<dt>m</dt>
<dd>Number of imputations</dd>
<dt>nmis</dt>
<dd>Named integer vector of missing counts per variable</dd>
<dt>method</dt>
<dd>Named list of methods used per variable</dd>
<dt>boot</dt>
<dd>Logical: was bootstrap used?</dd>
<dt>uncert</dt>
<dd>Character: uncertainty method used</dd>
<dt>call</dt>
<dd>The original function call</dd>
<dt>tuning_log</dt>
<dd>Optional tuning report from the (single) tuning run,
shared by all m imputations; NULL when tuning was not requested</dd>
<dt>chain</dt>
<dd>Optional per-iteration chain statistics:
<code>list(mean = , var = )</code> of arrays with dimensions
<code>[variable, iteration, imputation]</code> holding the mean and variance of
the imputed values of each numeric variable after each sequential
iteration (the basis of <code>plot.vimmi</code> trace plots)</dd>
<dt>seed</dt>
<dd>The seed applied at entry of the <code>vimpute()</code> call
(NULL when none was set)</dd>
<dt>model_error</dt>
<dd>Optional per-variable model-quality list (NRMSE/PFC,
from the first imputation run), as in <code>attr(result, "model_error")</code>
of single runs</dd></dl>

Arguments

Constructor for vimmi objects — new_vimmi

<dl>

<dt>data</dt>
<dd>Original data.frame/data.table with NAs intact</dd>


<dt>imp</dt>
<dd>Named list: per variable with missings, a data.frame with
nmis rows and m columns of imputed values</dd>


<dt>where</dt>
<dd>Logical matrix indicating which cells were imputed</dd>


<dt>m</dt>
<dd>Number of imputations</dd>


<dt>nmis</dt>
<dd>Named integer vector of missing counts per variable</dd>


<dt>method</dt>
<dd>Named list of methods used per variable</dd>


<dt>boot</dt>
<dd>Logical: was bootstrap used?</dd>


<dt>uncert</dt>
<dd>Character: uncertainty method used</dd>


<dt>call</dt>
<dd>The original function call</dd>


<dt>tuning_log</dt>
<dd>Optional tuning report from the (single) tuning run,
shared by all m imputations; NULL when tuning was not requested</dd>


<dt>chain</dt>
<dd>Optional per-iteration chain statistics:
<code>list(mean = , var = )</code> of arrays with dimensions
<code>[variable, iteration, imputation]</code> holding the mean and variance of
the imputed values of each numeric variable after each sequential
iteration (the basis of <code>plot.vimmi</code> trace plots)</dd>


<dt>seed</dt>
<dd>The seed applied at entry of the <code>vimpute()</code> call
(NULL when none was set)</dd>


<dt>model_error</dt>
<dd>Optional per-variable model-quality list (NRMSE/PFC,
from the first imputation run), as in <code>attr(result, "model_error")</code>
of single runs</dd>

</dl>

Constructor for vimmi objects

new_vimmi: Constructor for vimmi objects

Description

Usage

Value

Arguments