clean.cp

(Over-)estimate and remove change-points in the mean for scale estimation and bootstrapping

internal

Binary segmentation methods for detecting and estimating multiple change-points in the mean or second-order structure of high-dimensional time series as described in Cho and Fryzlewicz (2014) <doi:10.1111/rssb.12079> and Cho (2016) <doi:10.1214/16-EJS1155>.

Haeran Cho

hdbinseg

Change-Point Analysis of High-Dimensional Time Series via Binary
Segmentation

Piotr Fryzlewicz

clean.cp function

<dl><dt>z</dt>
<dd>input data matrix, with each row representing the component time series</dd>
<dt>type</dt>
<dd>if <code>type = 'dcbs'</code>, a binary tree of given <code>height</code> is grown using DCBS algorithm without thresholding, if <code>type = 'sbs'</code>, the binary trees is grown using the SBS algorithm with thresholds chosen small</dd>
<dt>phi, trim, height</dt>
<dd>see <code>dcbs.alg</code></dd></dl>

Arguments

Removing change-points in the mean — clean.cp

<dl>

<dt>z</dt>
<dd>input data matrix, with each row representing the component time series</dd>


<dt>type</dt>
<dd>if <code>type = 'dcbs'</code>, a binary tree of given <code>height</code> is grown using DCBS algorithm without thresholding, if <code>type = 'sbs'</code>, the binary trees is grown using the SBS algorithm with thresholds chosen small</dd>


<dt>phi, trim, height</dt>
<dd>see <code>dcbs.alg</code></dd>

</dl>

clean.cp: Removing change-points in the mean

Description

Usage

Value

Arguments