Multiscale change-point segmentation: beyond step functions
Authors Li H, Guo Q, Munk A Journal Electronic Journal of Statistics Citation Electron. J. Statist. Volume 13, Number 2 (2019), 3254-3296. Abstract Modern multiscale type segmentation methods are known to detect multiple change-points with high statistical accuracy, while allowing for fast computation. Underpinning (minimax) estimation theory has been developed