not: Narrowest-Over-Threshold Change-Point Detection

Provides efficient implementation of the Narrowest-Over-Threshold methodology for detecting an unknown number of change-points occurring at unknown locations in one-dimensional data following deterministic signal + noise model, see R. Baranowski, Y. Chen and P. Fryzlewicz (2019) <doi:10.1111/rssb.12322>. Currently implemented scenarios are: piecewise-constant signal, piecewise-constant signal with a heavy-tailed noise, piecewise-linear signal, piecewise-quadratic signal, piecewise-constant signal and with piecewise-constant variance of the noise.

Package details

AuthorRafal Baranowski, Yining Chen, Piotr Fryzlewicz
MaintainerRafal Baranowski <>
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:

Try the not package in your browser

Any scripts or data that you put into this service are public.

not documentation built on July 5, 2019, 1:02 a.m.