diff_vecs

Computes the sum of absolute differences in the finite non-<code>NA</code>/<code>NULL</code> elements between two vectors.

Implementation of the Generalized Score Matching estimator in Yu et al. (2019) <https://jmlr.org/papers/v20/18-278.html> for non-negative graphical models (truncated Gaussian, exponential square-root, gamma, a-b models) and univariate truncated Gaussian distributions. Also includes the original estimator for untruncated Gaussian graphical models from Lin et al. (2016) <doi:10.1214/16-EJS1126>, with the addition of a diagonal multiplier.

Shiqing Yu

genscore

Generalized Score Matching Estimators

diff_vecs function

<dl><dt>l1</dt>
<dd>A vector.</dd>
<dt>l2</dt>
<dd>A vector.</dd>
<dt>relative</dt>
<dd>A boolean, default to <code>FALSE</code>. If <code>TRUE</code>, returns the relative difference (sum of absolute differences divided by the elementwise minimum between <code>l1</code> and <code>l2</code>).</dd></dl>

Arguments

Computes the sum of absolute differences in the finite non-NA/NULL elements between two vectors. — diff_vecs

<dl>

<dt>l1</dt>
<dd>A vector.</dd>


<dt>l2</dt>
<dd>A vector.</dd>


<dt>relative</dt>
<dd>A boolean, default to <code>FALSE</code>. If <code>TRUE</code>, returns the relative difference (sum of absolute differences divided by the elementwise minimum between <code>l1</code> and <code>l2</code>).</dd>

</dl>

diff_vecs: Computes the sum of absolute differences in the finite non-NA/NULL elements between two vectors.

Description

Usage

Value

Arguments