predict.abess

Make predictions from a fitted "<code>abess</code>" object.

Extremely efficient toolkit for solving the best subset selection problem <https://www.jmlr.org/papers/v23/21-1060.html>. This package is its R interface. The package implements and generalizes algorithms designed in <doi:10.1073/pnas.2014241117> that exploits a novel sequencing-and-splicing technique to guarantee exact support recovery and globally optimal solution in polynomial times for linear model. It also supports best subset selection for logistic regression, Poisson regression, Cox proportional hazard model, Gamma regression, multiple-response regression, multinomial logistic regression, ordinal regression, (sequential) principal component analysis, and robust principal component analysis. The other valuable features such as the best subset of group selection <doi:10.1287/ijoc.2022.1241> and sure independence screening <doi:10.1111/j.1467-9868.2008.00674.x> are also provided.

Jin Zhu

abess

Fast Best Subset Selection

Zezhi Wang

Liyuan Hu

Junhao Huang

Kangkang Jiang

Yanhang Zhang

Borui Tang

Shiyun Lin

Junxian Zhu

Canhong Wen

Heping Zhang

Xueqin Wang

spectra contributors 

predict.abess function

<dl><dt>object</dt>
<dd>An "<code>abess</code>" project.</dd>
<dt>newx</dt>
<dd>New data used for prediction. If omitted, the fitted linear predictors are used.</dd>
<dt>type</dt>
<dd><code>type = "link"</code> gives the linear predictors for <code>"binomial"</code>,
<code>"poisson"</code> or <code>"cox"</code> models; for <code>"gaussian"</code> models it gives the
fitted values. <code>type = "response"</code> gives the fitted probabilities for
<code>"binomial"</code> and <code>"ordinal"</code>, fitted mean for <code>"poisson"</code> and the fitted relative-risk for
<code>"cox"</code>; for <code>"gaussian"</code>, <code>type = "response"</code> is equivalent to <code>type = "link"</code>.</dd>
<dt>support.size</dt>
<dd>An integer value specifies
the model size fitted at given <code>support.size</code>.
If <code>support.size = NULL</code>, then the model with
the best tuning value would be returned.
Default: <code>support.size = NULL</code>.</dd>
<dt>...</dt>
<dd>Additional arguments affecting the predictions produced.</dd></dl>

Arguments

Make predictions from a fitted "abess" object. — predict.abess

<dl>

<dt>object</dt>
<dd>An "<code>abess</code>" project.</dd>


<dt>newx</dt>
<dd>New data used for prediction. If omitted, the fitted linear predictors are used.</dd>


<dt>type</dt>
<dd><code>type = "link"</code> gives the linear predictors for <code>"binomial"</code>,
<code>"poisson"</code> or <code>"cox"</code> models; for <code>"gaussian"</code> models it gives the
fitted values. <code>type = "response"</code> gives the fitted probabilities for
<code>"binomial"</code> and <code>"ordinal"</code>, fitted mean for <code>"poisson"</code> and the fitted relative-risk for
<code>"cox"</code>; for <code>"gaussian"</code>, <code>type = "response"</code> is equivalent to <code>type = "link"</code>.</dd>


<dt>support.size</dt>
<dd>An integer value specifies
the model size fitted at given <code>support.size</code>.
If <code>support.size = NULL</code>, then the model with
the best tuning value would be returned.
Default: <code>support.size = NULL</code>.</dd>


<dt>...</dt>
<dd>Additional arguments affecting the predictions produced.</dd>

</dl>

Make predictions from a fitted "<code>abess</code>" object.

predict.abess: Make predictions from a fitted "`abess`" object.

Description

Usage

Value

Arguments

See Also