stability.path-class

print,stability.path-method

show,stability.path-method

Class of object returned by the <code>stability</code> function, with
methods <code>print</code>, <code>show</code> and <code>plot</code>.

class

Fits classical sparse regression models with
efficient active set algorithms by solving quadratic problems as described by
Grandvalet, Chiquet and Ambroise (2017) <doi:10.48550/arXiv.1210.2077>. Also provides a few
methods for model selection purpose (cross-validation, stability selection).

Julien Chiquet

quadrupen

Sparsity by Worst-Case Quadratic Penalties

stability.path-class function

<dl><dt><code>probabilities</code>: </dt>
<dd>a <code>Matrix</code> object containing the
estimated probabilities of selection along the path of solutions.</dd><dt><code>penalty</code>: </dt>
<dd>Object of class <code>"character"</code>
indicating the penalizer used.</dd><dt><code>naive</code>: </dt>
<dd>logical indicating whether rescaling of the
coefficients has been performed regarding the \(\ell_2\)-penalty.</dd><dt><code>lambda1</code>: </dt>
<dd>a vector with the levels of the first penalty.</dd><dt><code>lambda2</code>: </dt>
<dd>a scalar with the \(\ell_2\)-penalty level.</dd><dt><code>folds</code>: </dt>
<dd>a list that contains the folds used for each subsample.</dd>
</dl>

Slots

Class "stability.path" — stability.path-class

<dl>

<dt><code>probabilities</code>: </dt>
<dd>a <code>Matrix</code> object containing the
estimated probabilities of selection along the path of solutions.</dd>

<dt><code>penalty</code>: </dt>
<dd>Object of class <code>"character"</code>
indicating the penalizer used.</dd>

<dt><code>naive</code>: </dt>
<dd>logical indicating whether rescaling of the
coefficients has been performed regarding the \(\ell_2\)-penalty.</dd>

<dt><code>lambda1</code>: </dt>
<dd>a vector with the levels of the first penalty.</dd>

<dt><code>lambda2</code>: </dt>
<dd>a scalar with the \(\ell_2\)-penalty level.</dd>

<dt><code>folds</code>: </dt>
<dd>a list that contains the folds used for each subsample.</dd>


</dl>

stability.path-class: Class "stability.path"

Description

Arguments

Slots

See Also