stabilise

Function to calculate stability of variables' association with an outcome for a given model over a number of bootstrap repeats

A stable approach to variable selection through stability selection and the use of a permutation-based objective stability threshold. Lima et al (2021) <doi:10.1038/s41598-020-79317-8>, Meinshausen and Buhlmann (2010) <doi:10.1111/j.1467-9868.2010.00740.x>.

Robert Hyde

stabiliser

Stabilising Variable Selection

Martin Green

Eliana Lima

stabilise function

<dl><dt>data</dt>
<dd>A dataframe containing an outcome variable to be permuted.</dd>
<dt>outcome</dt>
<dd>The outcome as a string (i.e. "y").</dd>
<dt>boot_reps</dt>
<dd>The number of bootstrap samples. Default is "auto" which selects number based on dataframe size.</dd>
<dt>permutations</dt>
<dd>The number of times to be permuted per repeat. Default is "auto" which selects number based on dataframe size.</dd>
<dt>perm_boot_reps</dt>
<dd>The number of times to repeat each set of permutations. Default is 20.</dd>
<dt>models</dt>
<dd>The models to select for stabilising. Default is elastic net (models = c("enet")), other available models include "lasso", "mbic", "mcp".</dd>
<dt>type</dt>
<dd>The type of model, either "linear" or "logistic"</dd>
<dt>quantile</dt>
<dd>The quantile of null stabilities to use as a threshold.</dd>
<dt>normalise</dt>
<dd>Normalise numeric variables (TRUE/FALSE)</dd>
<dt>dummy</dt>
<dd>Create dummy variables for factors/characters (TRUE/FALSE)</dd>
<dt>impute</dt>
<dd>Impute missing data (TRUE/FALSE)</dd></dl>

Arguments

stabilise — stabilise

<dl>

<dt>data</dt>
<dd>A dataframe containing an outcome variable to be permuted.</dd>


<dt>outcome</dt>
<dd>The outcome as a string (i.e. "y").</dd>


<dt>boot_reps</dt>
<dd>The number of bootstrap samples. Default is "auto" which selects number based on dataframe size.</dd>


<dt>permutations</dt>
<dd>The number of times to be permuted per repeat. Default is "auto" which selects number based on dataframe size.</dd>


<dt>perm_boot_reps</dt>
<dd>The number of times to repeat each set of permutations. Default is 20.</dd>


<dt>models</dt>
<dd>The models to select for stabilising. Default is elastic net (models = c("enet")), other available models include "lasso", "mbic", "mcp".</dd>


<dt>type</dt>
<dd>The type of model, either "linear" or "logistic"</dd>


<dt>quantile</dt>
<dd>The quantile of null stabilities to use as a threshold.</dd>


<dt>normalise</dt>
<dd>Normalise numeric variables (TRUE/FALSE)</dd>


<dt>dummy</dt>
<dd>Create dummy variables for factors/characters (TRUE/FALSE)</dd>


<dt>impute</dt>
<dd>Impute missing data (TRUE/FALSE)</dd>

</dl>

stabilise: stabilise

Description

Value

Arguments