deviances

Calculate deviances at provided coefficients

Efficient algorithm for solving PU (Positive and Unlabeled) problem in low or high dimensional setting with lasso or group lasso penalty. The algorithm uses Maximization-Minorization and (block) coordinate descent. Sparse calculation and parallel computing are supported for the computational speed-up. See Hyebin Song, Garvesh Raskutti (2018) <arXiv:1711.08129>.

Hyebin Song

PUlasso

High-Dimensional Variable Selection with Presence-Only Data

Garvesh Raskutti

deviances function

<dl><dt>X</dt>
<dd>Input matrix</dd>
<dt>z</dt>
<dd>Response vector</dd>
<dt>py1</dt>
<dd>True prevalence Pr(Y=1)</dd>
<dt>coefMat</dt>
<dd>A coefficient matrix whose column corresponds to a set of coefficients</dd>
<dt>weights</dt>
<dd>observation weights. Default is 1 for each observation.</dd></dl>

Arguments

Deviance — deviances

<dl>

<dt>X</dt>
<dd>Input matrix</dd>


<dt>z</dt>
<dd>Response vector</dd>


<dt>py1</dt>
<dd>True prevalence Pr(Y=1)</dd>


<dt>coefMat</dt>
<dd>A coefficient matrix whose column corresponds to a set of coefficients</dd>


<dt>weights</dt>
<dd>observation weights. Default is 1 for each observation.</dd>

</dl>

deviances: Deviance

Description

Usage

Value

Arguments

Examples