fastcpd: Fast Change Point Detection via Sequential Gradient Descent

Implements fast change point detection algorithm based on the paper "Sequential Gradient Descent and Quasi-Newton's Method for Change-Point Analysis" by Xianyang Zhang, Trisha Dawn <https://proceedings.mlr.press/v206/zhang23b.html>. The algorithm is based on dynamic programming with pruning and sequential gradient descent. See Li and Zhang (2026) <doi:10.18637/jss.v116.i06> for details.

Version: 1.0.0
Depends: R (≥ 2.10)
Imports: glmnet, Matrix, methods, Rcpp (≥ 0.11.0), stats
LinkingTo: progress, Rcpp, RcppArmadillo, testthat
Suggests: ggplot2, gridExtra, knitr, matrixStats, mvtnorm, RcppArmadillo, rmarkdown, testthat (≥ 3.0.0), xml2
Published: 2026-06-16
DOI: 10.32614/CRAN.package.fastcpd
Author: Xingchi Li ORCID iD [aut, cre, cph], Xianyang Zhang [aut, cph]
Maintainer: Xingchi Li <anthony.li.stat.tamu.edu at lixingchi.com>
BugReports: https://github.com/doccstat/fastcpd/issues
License: Apache License (≥ 2)
URL: https://fastcpd.xingchi.li, https://github.com/doccstat/fastcpd
NeedsCompilation: yes
Citation: fastcpd citation info
Materials: README, NEWS
In views: TimeSeries
CRAN checks: fastcpd results

Documentation:

Reference manual: fastcpd.html , fastcpd.pdf
Vignettes: Comparison with other R packages (source)
Comparison with vanilla PELT (source)
Advanced examples (source)
Custom logistic regression model (source)
Exploration during development (source)
Time Complexity Analysis in fastcpd (source)

Downloads:

Package source: fastcpd_1.0.0.tar.gz
Windows binaries: r-devel: fastcpd_0.16.2.zip, r-release: fastcpd_1.0.0.zip, r-oldrel: fastcpd_0.16.2.zip
macOS binaries: r-release (arm64): fastcpd_1.0.0.tgz, r-oldrel (arm64): fastcpd_1.0.0.tgz, r-release (x86_64): fastcpd_1.0.0.tgz, r-oldrel (x86_64): fastcpd_1.0.0.tgz
Old sources: fastcpd archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=fastcpd to link to this page.