mscale

Compute the M-scale without centering the values.

Robust penalized (adaptive) elastic net S and M estimators for
linear regression. The adaptive methods are proposed in
Kepplinger, D. (2023) <doi:10.1016/j.csda.2023.107730> and the
non-adaptive methods in
Cohen Freue, G. V., Kepplinger, D., Salibián-Barrera, M., and Smucler, E.
(2019) <doi:10.1214/19-AOAS1269>.
The package implements robust hyper-parameter selection with robust
information sharing cross-validation according to Kepplinger & Wei (2025)
<doi:10.1080/00401706.2025.2540970>.

David Kepplinger

pense

Penalized Elastic Net S/MM-Estimator of Regression

Matías Salibián-Barrera

Gabriela Cohen Freue

mscale function

<dl><dt>x</dt>
<dd>numeric values. Missing values are verbosely ignored.</dd>
<dt>bdp</dt>
<dd>desired breakdown point (between 0 and 0.5).</dd>
<dt>cc</dt>
<dd>tuning parameters for the chosen rho function.
By default, chosen to yield a consistent estimate for the Normal distribution.</dd>
<dt>opts</dt>
<dd>a list of options for the M-scale estimation algorithm,
see <code>mscale_algorithm_options()</code> for details.</dd></dl>

Arguments

Compute the M-Scale of Centered Values — mscale

<dl>

<dt>x</dt>
<dd>numeric values. Missing values are verbosely ignored.</dd>


<dt>bdp</dt>
<dd>desired breakdown point (between 0 and 0.5).</dd>


<dt>cc</dt>
<dd>tuning parameters for the chosen rho function.
By default, chosen to yield a consistent estimate for the Normal distribution.</dd>


<dt>opts</dt>
<dd>a list of options for the M-scale estimation algorithm,
see <code>mscale_algorithm_options()</code> for details.</dd>

</dl>

mscale: Compute the M-Scale of Centered Values

Description

Usage

Value

Arguments

See Also