Package: ocd 1.1

Yudong Chen

ocd: High-Dimensional Multiscale Online Changepoint Detection

Implements the algorithm in Chen, Wang and Samworth (2020) <arxiv:2003.03668> for online detection of sudden mean changes in a sequence of high-dimensional observations. It also implements methods by Mei (2010) <doi:10.1093/biomet/asq010>, Xie and Siegmund (2013) <doi:10.1214/13-AOS1094> and Chan (2017) <doi:10.1214/17-AOS1546>.

Authors:Yudong Chen, Tengyao Wang, Richard J. Samworth

ocd_1.1.tar.gz
ocd_1.1.tar.gz(r-4.5-noble)ocd_1.1.tar.gz(r-4.4-noble)
ocd_1.1.tgz(r-4.4-emscripten)ocd_1.1.tgz(r-4.3-emscripten)
ocd.pdf |ocd.html
ocd/json (API)

# Install 'ocd' in R:
install.packages('ocd', repos = 'https://cloud.r-project.org')
Datasets:

On CRAN:

Conda:

This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.

1.70 score 153 downloads 31 exports 0 dependencies

Last updated 4 years agofrom:b21cde5960. Checks:1 OK, 2 NOTE. Indexed: yes.

TargetResultLatest binary
Doc / VignettesOKMar 31 2025
R-4.5-linuxNOTEMar 31 2025
R-4.4-linuxNOTEMar 31 2025

Exports:baselineMeanbaselineSDChan_updateChangepointDetectorcheckChangedata_dimgetDataMC_ChanMC_MeiMC_ocdMC_XSMei_updaten_obsnew_Channew_Meinew_OCDnew_XSnormalisedStatisticsocd_updateocdMethodparampatienceresetsetBaselineMeansetBaselineSDsetStatusstatisticsstatusthresholdstrackedXS_update

Dependencies:

Citation

To cite package ‘ocd’ in publications use:

Chen Y, Wang T, Samworth RJ (2020). ocd: High-Dimensional Multiscale Online Changepoint Detection. R package version 1.1, https://CRAN.R-project.org/package=ocd.

ATTENTION: This citation information has been auto-generated from the package DESCRIPTION file and may need manual editing, see ‘help("citation")’.

Corresponding BibTeX entry:

  @Manual{,
    title = {ocd: High-Dimensional Multiscale Online Changepoint
      Detection},
    author = {Yudong Chen and Tengyao Wang and Richard J. Samworth},
    year = {2020},
    note = {R package version 1.1},
    url = {https://CRAN.R-project.org/package=ocd},
  }

Readme and manuals

ocd: A package for high-dimensional multiscale online changepoint detection

Description

The ocd R package provides the S3 class ChangepointDetector that processes data sequentially and aims to detect change as soon as it occurs online subject to false alarm rates.

Installation

library(devtools)
install_github('wangtengyao/ocd')

References

Chen, Y., Wang, T. and Samworth, R. J. (2020) High-dimensional multiscale online changepoint detection Preprint. arxiv:2003.03668.

Help Manual

Help pageTopics
Accessor functions to attributes of class ChangepointDetectoraccessor baselineMean baselineSD data_dim n_obs ocdMethod param patience statistics status thresholds tracked
Processing a new data point for the 'Chan' classChan_update
Constructor for the ChangepointDetector S3 classChangepointDetector
Check if a mean change has occurred.checkChange
Processing a new data pointgetData getData.Chan getData.Mei getData.OCD getData.XS
Compute Monte Carlo thresholds for the Chan methodMC_Chan
Compute Monte Carlo thresholds for the Mei methodMC_Mei
Compute Monte Carlo thresholds for the OCD methodMC_ocd
Compute Monte Carlo thresholds for the XS methodMC_XS
Processing a new data point for the 'Mei' classMei_update
construtor for subclass 'Chan' in class 'ChangepointDetector'new_Chan
constructor of subclass 'Mei' in class 'ChangepointDetector'new_Mei
constructor of subclass 'OCD' in class 'ChangepointDetector'new_OCD
constructor of subclass 'XS' in class 'ChangepointDetector'new_XS
Compute maximum ratio between detection statistic and its thresholdnormalisedStatistics
ocd: A package for high-dimensional multiscale online changepoint detectionocd-package ocd
Processing a new data point for the 'OCD' classocd_update
Parkfield seismic sensor dataParkfieldSensors
Printing methods for the 'ChangepointDetector' classprint.ChangepointDetector
Reset changepoint detector to initial statereset reset.Chan reset.Mei reset.OCD reset.XS
Set baseline meansetBaselineMean
Set baseline standard deviationsetBaselineSD
Set changepoint detector statussetStatus
compute new mean and sd from old ones with one additional observationupdate_param
Processing a new data point for the 'XS' classXS_update