not-package {not}

Narrowest-Over-Threshold Change-Point Detection

Description

Implements the Narrowest-Over-Threshold approach for general multiple change-point detection in one-dimensional data following 'deterministic signal + noise' model. Scenarios that are currently implemented 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 standard deviation of the noise. The main routines of the package are not and features.

References

R. Baranowski, Y. Chen, and P. Fryzlewicz (2019). Narrowest-Over-Threshold Change-Point Detection. (http://stats.lse.ac.uk/fryzlewicz/not/not.pdf)