Wild binary segmentation for multiple change-point detection
Wild binary segmentation for multiple change-point detection
About this item
Full title
Author / Creator
Publisher
Ithaca: Cornell University Library, arXiv.org
Journal title
Language
English
Formats
Publication information
Publisher
Ithaca: Cornell University Library, arXiv.org
Subjects
More information
Scope and Contents
Contents
We propose a new technique, called wild binary segmentation (WBS), for consistent estimation of the number and locations of multiple change-points in data. We assume that the number of change-points can increase to infinity with the sample size. Due to a certain random localisation mechanism, WBS works even for very short spacings between the chang...
Alternative Titles
Full title
Wild binary segmentation for multiple change-point detection
Authors, Artists and Contributors
Author / Creator
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_proquest_journals_2084628129
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_2084628129
Other Identifiers
E-ISSN
2331-8422
DOI
10.48550/arxiv.1411.0858
