lcv_kdens_symmat

Given a cube of sample observations (consisting of random symmetric positive definite matrices), and a vector of candidate bandwidth parameters <code>b</code>,
compute the leave-one-out likelihood cross-validation criterion and
return the bandwidth among the choices that minimizes the criterion.

Kernel smoothing for Wishart random matrices described in Daayeb, Khardani and Ouimet (2025) <doi:10.48550/arXiv.2506.08816>, Gaussian and log-Gaussian models using least square or likelihood cross validation criteria for optimal bandwidth selection.

Leo Belzile

Kernel Density Estimation for Random Symmetric Positive Definite
Matrices

Frederic Ouimet

lcv_kdens_symmat function

<dl><dt>x</dt>
<dd>array of dimension <code>d</code> by <code>d</code> by <code>n</code></dd>
<dt>b</dt>
<dd>vector of candidate bandwidth, strictly positive</dd>
<dt>kernel</dt>
<dd>string indicating the kernel, one of <code>Wishart</code> or <code>smlnorm</code>.</dd></dl>

Arguments

Likelihood cross-validation for symmetric positive definite matrix kernels — lcv_kdens_symmat

<dl>

<dt>x</dt>
<dd>array of dimension <code>d</code> by <code>d</code> by <code>n</code></dd>


<dt>b</dt>
<dd>vector of candidate bandwidth, strictly positive</dd>


<dt>kernel</dt>
<dd>string indicating the kernel, one of <code>Wishart</code> or <code>smlnorm</code>.</dd>

</dl>

lcv_kdens_symmat: Likelihood cross-validation for symmetric positive definite matrix kernels

Description

Usage

Value

Arguments