multiscale.grid

Create asymmetric bandwidth grids to be used with <code>multiscale.localPrune</code>

internal

Implementations of MOSUM-based statistical procedures and algorithms for detecting multiple changes in the mean. This comprises the MOSUM procedure for estimating multiple mean changes from Eichinger and Kirch (2018) <doi:10.3150/16-BEJ887> and the multiscale algorithmic extension from Cho and Kirch (2022) <doi:10.1007/s10463-021-00811-5>, as well as the bootstrap procedure for generating confidence intervals about the locations of change points as proposed in Cho and Kirch (2022) <doi:10.1016/j.csda.2022.107552>. See also Meier, Kirch and Cho (2021) <doi:10.18637/jss.v097.i08> which accompanies the R package.

Haeran Cho

mosum

Moving Sum Based Procedures for Changes in the Mean

Alexander Meier

Claudia Kirch

multiscale.grid function

<dl><dt>bandwidths.left</dt>
<dd>left parts of the bandwidths</dd>
<dt>bandwidths.right</dt>
<dd>right parts of the bandwidths</dd>
<dt>method</dt>
<dd>how the asymmetric bandwidths are created;
possible values are<ul>
<li><code>'cartesian'</code>create all bandwidths in the Cartesian product of 
 bandwidths.left and bandwidths.right</li>
<li><code>'concatenate'</code>join bandwidths.left and bandwidths.right element-wise</li>
</ul></dd>
<dt>max.unbalance</dt>
<dd>a numeric value for the maximal ratio between maximal and minimal bandwidth,
<code>1 &lt;= max.unbalance &lt;= Inf</code>; use iff <code>method='cartesian'</code></dd></dl>

Arguments

Multiscale bandwidth grids — multiscale.grid

<dl>

<dt>bandwidths.left</dt>
<dd>left parts of the bandwidths</dd>


<dt>bandwidths.right</dt>
<dd>right parts of the bandwidths</dd>


<dt>method</dt>
<dd>how the asymmetric bandwidths are created;
possible values are<ul>
<li><code>'cartesian'</code>create all bandwidths in the Cartesian product of 
 bandwidths.left and bandwidths.right</li>
<li><code>'concatenate'</code>join bandwidths.left and bandwidths.right element-wise</li>
</ul></dd>


<dt>max.unbalance</dt>
<dd>a numeric value for the maximal ratio between maximal and minimal bandwidth,
<code>1 &lt;= max.unbalance &lt;= Inf</code>; use iff <code>method='cartesian'</code></dd>

</dl>

multiscale.grid: Multiscale bandwidth grids

Description

Usage

Value

Arguments