dantzigS

The Dantzig selector (DS) finds a solution for the model parameters
of a linear model, <code>beta</code> using linear programming. For a given <code>delta</code>,
DS minimizes the L_1-norm (sum of absolute values) of <code>beta</code> subject to the constraint
that <code>max(|t(X)(y-X * beta)|)</code>&lt;= <code>delta</code>.

internal

The method aims to identify important factors in screening experiments by aggregation over random models as studied in Singh and Stufken (2022) <doi:10.48550/arXiv.2205.13497>. This package provides functions to run the Gauss-Dantzig selector on screening experiments when interactions may be affecting the response. Currently, all functions require each factor to be at two levels coded as +1 and -1.

Rakhi Singh

GDSARM

Gauss - Dantzig Selector: Aggregation over Random Models

John Stufken

dantzigS function

<dl><dt>X</dt>
<dd>a design matrix.</dd>
<dt>y</dt>
<dd>a vector of responses.</dd>
<dt>delta</dt>
<dd>the specific value of <code>delta</code> for which the Dantzig Selector
optimization needs to be solved</dd>
<dt>scale.X</dt>
<dd>a number by which each column of <code>X</code> should be scaled</dd></dl>

Arguments

Dantzig selector using the lpsolve package — dantzigS

<dl>

<dt>X</dt>
<dd>a design matrix.</dd>


<dt>y</dt>
<dd>a vector of responses.</dd>


<dt>delta</dt>
<dd>the specific value of <code>delta</code> for which the Dantzig Selector
optimization needs to be solved</dd>


<dt>scale.X</dt>
<dd>a number by which each column of <code>X</code> should be scaled</dd>

</dl>

dantzigS: Dantzig selector using the lpsolve package

Description

Usage

Value

Arguments