gen.input

Generate Haar wavelet transformation of time series containing information about possible change-points in the second-order structure of time series

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

gen.input function

<dl><dt>x</dt>
<dd>input data matrix, with each row representing the component time series</dd>
<dt>scales</dt>
<dd>negative integers for wavelet scales, with a small negative integer representing a fine scale</dd>
<dt>sq</dt>
<dd>if <code>sq = TRUE</code>, squared root of wavelet periodograms are used for change-point analysis</dd>
<dt>diag</dt>
<dd>if <code>diag = TRUE</code>, only changes in the diagonal elements of the autocovariance matrices are searched for</dd>
<dt>sgn</dt>
<dd>if <code>diag = FALSE</code>, wavelet transformations of the cross-covariances are computed with the matching signs</dd></dl>

Arguments

Generating Haar wavelet transformation of time series — gen.input

<dl>

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


<dt>scales</dt>
<dd>negative integers for wavelet scales, with a small negative integer representing a fine scale</dd>


<dt>sq</dt>
<dd>if <code>sq = TRUE</code>, squared root of wavelet periodograms are used for change-point analysis</dd>


<dt>diag</dt>
<dd>if <code>diag = TRUE</code>, only changes in the diagonal elements of the autocovariance matrices are searched for</dd>


<dt>sgn</dt>
<dd>if <code>diag = FALSE</code>, wavelet transformations of the cross-covariances are computed with the matching signs</dd>

</dl>

gen.input: Generating Haar wavelet transformation of time series

Description

Usage

Value

Arguments