Multiscale Change-point Segmentation: Beyond Step Functions

Abstract

Modern multiscale type segmentation methods are known to detect multiple change-points with high statistical accuracy, while allowing for fast computation. Underpinning theory has been developed mainly for models that assume the signal as a piecewise constant function. In this paper this will be extended to certain function classes beyond such step functions in a nonparametric regression setting, revealing certain multiscale segmentation methods as robust to deviation from such piecewise constant functions. Our main finding is the adaptation over such function classes for a universal thresholding, which includes bounded variation functions, and (piecewise) Hölder functions of smoothness order $0<\alpha \le 1$ as special cases. From this we derive statistical guarantees on feature detection in terms of jumps and modes. Another key finding is that these multiscale segmentation methods perform nearly (up to a log-factor) as well as the oracle piecewise constant segmentation estimator (with known jump locations), and the best piecewise constant approximants of the (unknown) true signal. Theoretical findings are examined by various numerical simulations.

References in corpus (13)

• Wild binary segmentation for multiple change-point detection
• Consistencies and rates of convergence of jump-penalized least squares estimators
• Optimal and fast detection of spatial clusters with scan statistics
• Multiple testing of local maxima for detection of peaks in 1D
• FDR-Control in Multiscale Change-point Segmentation
• Extreme-Value Analysis of Standardized Gaussian Increments
• Approximate Recovery in Changepoint Problems, from ${\ell }_2$ Estimation Error Rates
• Identifiability for Blind Source Separation of Multiple Finite Alphabet Linear Mixtures
• Multi-sequence segmentation via score and higher-criticism tests
• Variational Multiscale Nonparametric Regression: Smooth Functions
• Multiple Testing of Local Extrema for Detection of Change Points
• Frame-constrained Total Variation Regularization for White Noise Regression
• Segmentation and Estimation of Change-point Models: False Positive Control and Confidence Regions

Cited by in corpus (7)

• Seeded Binary Segmentation: A general methodology for fast and optimal change point detection
• Data segmentation algorithms: Univariate mean change and beyond
• A review on minimax rates in change point detection and localisation
• Univariate Mean Change Point Detection: Penalization, CUSUM and Optimality
• A Variational View on Statistical Multiscale Estimation
• Frame-constrained Total Variation Regularization for White Noise Regression
• Robust inference of cooperative behaviour of multiple ion channels in voltage-clamp recordings