cd_algorithm_options

Set options for the CD algorithm to compute adaptive EN S-estimates.

Robust penalized (adaptive) elastic net S and M estimators for
linear regression. The methods are proposed in
Cohen Freue, G. V., Kepplinger, D., Salibián-Barrera, M., and Smucler, E.
(2019) <https://projecteuclid.org/euclid.aoas/1574910036>.
The package implements the extensions and algorithms described in
Kepplinger, D. (2020) <doi:10.14288/1.0392915>.

David Kepplinger

pense

Penalized Elastic Net S/MM-Estimator of Regression

Matías Salibián-Barrera

Gabriela Cohen Freue

cd_algorithm_options function

<dl><dt>max_it</dt>
<dd>maximum number of iterations.</dd>
<dt>reset_it</dt>
<dd>number of iterations after which the residuals are
re-computed from scratch, to prevent numerical drifts from incremental
updates.</dd>
<dt>linesearch_steps</dt>
<dd>maximum number of steps used for line search.</dd>
<dt>linesearch_mult</dt>
<dd>multiplier to adjust the step size in the line
search.</dd></dl>

Arguments

Coordinate Descent (CD) Algorithm to Compute Penalized Elastic Net
S-estimates — cd_algorithm_options

<dl>

<dt>max_it</dt>
<dd>maximum number of iterations.</dd>


<dt>reset_it</dt>
<dd>number of iterations after which the residuals are
re-computed from scratch, to prevent numerical drifts from incremental
updates.</dd>


<dt>linesearch_steps</dt>
<dd>maximum number of steps used for line search.</dd>


<dt>linesearch_mult</dt>
<dd>multiplier to adjust the step size in the line
search.</dd>

</dl>

cd_algorithm_options: Coordinate Descent (CD) Algorithm to Compute Penalized Elastic Net S-estimates

Description

Usage

Value

Arguments

See Also