make_irls_warmup

This function may be used to create potentially valid starting
values for calling <code>sparsegl()</code> with a <code><a href="/link/stats%3A%3Afamily()?package=sparsegl&version=1.1.1" data-mini-rdoc="sparsegl::stats::family()">stats::family()</a></code> object.
It is not typically necessary to call this function (as it is used
internally to create some), but in some cases, especially with custom
generalized linear models, it may improve performance.

Efficient implementation of sparse group lasso with optional
bound constraints on the coefficients; see
<doi:10.18637/jss.v110.i06>. It supports the use of a sparse design
matrix as well as returning coefficient estimates in a sparse matrix.
Furthermore, it correctly calculates the degrees of freedom to allow
for information criteria rather than cross-validation with very large
data. Finally, the interface to compiled code avoids unnecessary
copies and allows for the use of long integers.

Daniel McDonald

sparsegl

Sparse Group Lasso

Daniel J. McDonald

Xiaoxuan Liang

Anibal Solón Heinsfeld

Aaron Cohen

Yi Yang

Hui Zou

Jerome Friedman

Trevor Hastie

Rob Tibshirani

Balasubramanian Narasimhan

Kenneth Tay

Noah Simon

Junyang Qian

James Yang

make_irls_warmup function

<dl><dt>nobs</dt>
<dd>Number of observations in the response (or rows in <code>x</code>).</dd>
<dt>nvars</dt>
<dd>Number of columns in <code>x</code></dd>
<dt>b0</dt>
<dd>Scalar. Initial value for the intercept.</dd>
<dt>beta</dt>
<dd>Vector. Initial values for the coefficients. Must be length
<code>nvars</code> (or a scalar).</dd>
<dt>r</dt>
<dd>Vector. Initial values for the deviance residuals. Must be length
<code>nobs</code> (or a scalar).</dd></dl>

Arguments

This function may be used to create potentially valid starting
values for calling <code>sparsegl()</code> with a <code><a href='https://rdrr.io/r/stats/family.html'>stats::family()</a></code> object.
It is not typically necessary to call this function (as it is used
internally to create some), but in some cases, especially with custom
generalized linear models, it may improve performance.

Create starting values for iterative reweighted least squares — make_irls_warmup

<dl>

<dt>nobs</dt>
<dd>Number of observations in the response (or rows in <code>x</code>).</dd>


<dt>nvars</dt>
<dd>Number of columns in <code>x</code></dd>


<dt>b0</dt>
<dd>Scalar. Initial value for the intercept.</dd>


<dt>beta</dt>
<dd>Vector. Initial values for the coefficients. Must be length
<code>nvars</code> (or a scalar).</dd>


<dt>r</dt>
<dd>Vector. Initial values for the deviance residuals. Must be length
<code>nobs</code> (or a scalar).</dd>

</dl>

Create starting values for iterative reweighted least squares

make_irls_warmup: Create starting values for iterative reweighted least squares

Description

Usage

Value

Arguments

Details