Consistencies and rates of convergence of jump-penalized least squares estimators

DOI10.1214/07-AOS558zbMATH Open1155.62034arXiv0902.4838MaRDI QIDQ1002154FDOQ1002154

Volkmar Liebscher, Olaf Wittich, Axel Munk, Leif Boysen, Angela Kempe

Publication date: 25 February 2009

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: We study the asymptotics for jump-penalized least squares regression aiming at approximating a regression function by piecewise constant functions. Besides conventional consistency and convergence rates of the estimates in $L^2([0,1))$ our results cover other metrics like Skorokhod metric on the space of c`{a}dl`{a}g functions and uniform metrics on $C([0,1])$. We will show that these estimators are in an adaptive sense rate optimal over certain classes of "approximation spaces." Special cases are the class of functions of bounded variation (piecewise) H"{o}lder continuous functions of order $0<alphale1$ and the class of step functions with a finite but arbitrary number of jumps. In the latter setting, we will also deduce the rates known from change-point analysis for detecting the jumps. Finally, the issue of fully automatic selection of the smoothing parameter is addressed.

Full work available at URL: https://arxiv.org/abs/0902.4838

Recommendations

• Scale space consistency of piecewise constant least squares estimators – another look at the regressogram
• Complexity penalized least squares estimators: Analytical results
• Jump estimation in inverse regression
• scientific article; zbMATH DE number 10785
• Nonparametric estimation of piecewise smooth regression functions

zbMATH Keywords

variable selectionpenalized maximum likelihoodadaptive estimationchange-point analysisjump detectionSkorokhod topologyPotts functionalapproximation spacesmultiscale resolution analysisregressogram

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Nonparametric regression and quantile regression (62G08) Asymptotic properties of nonparametric inference (62G20) Central limit and other weak theorems (60F05) Approximation by polynomials (41A10) Rate of convergence, degree of approximation (41A25)

Cites Work

• Local extremes, runs, strings and multiresolution. (With discussion)
• Ideal spatial adaptation by wavelet shrinkage
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Estimating the number of change-points via Schwarz' criterion
• Change point estimation using nonparametric regression
• Locally adaptive regression splines
• Discontinuous versus smooth regression
• Sparse estimators and the oracle property, or the return of Hodges' estimator
• Title not available (Why is that?)
• Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
• Title not available (Why is that?)
• On a Conjecture of Revesz
• Multiple changepoint fitting via quasilikelihood, with application to DNA sequence segmentation
• Scale space view of curve estimation.
• Title not available (Why is that?)
• For most large underdetermined systems of linear equations the minimal 𝓁₁‐norm solution is also the sparsest solution
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Minimal penalties for Gaussian model selection
• An introduction to \(\Gamma\)-convergence
• Title not available (Why is that?)
• Change-points in nonparametric regression analysis
• Estimation of a function with discontinuities via local polynomial fit with an adaptive window choice
• Inference about the change-point in a sequence of random variables
• Smoothers for Discontinuous Signals
• Don't shed tears over breaks
• Title not available (Why is that?)
• Wedgelets: Nearly minimax estimation of edges
• Scale space consistency of piecewise constant least squares estimators – another look at the regressogram
• For most large underdetermined systems of equations, the minimal 𝓁₁‐norm near‐solution approximates the sparsest near‐solution
• Edge-Preserving Smoothers for Image Processing
• CART and best-ortho-basis: a connection
• Image denoising: Pointwise adaptive approach
• Least squares estimation with complexity penalties
• Piecewise polynomial approximation on optimal meshes
• Epi-convergence of sequences of normal integrands and strong consistency of the maximum likelihood estimator
• Nonparametric estimation of piecewise smooth regression functions
• Robust priors for smoothing and image restoration
• A nonlinear gaussian filter applied to images with discontinuities
• On the law of the iterated logarithm for double sequences of random variables

Cited In (61)

• Efficient multiple change point detection for high‐dimensional generalized linear models
• Consistent change-point detection with kernels
• Recursive computation of piecewise constant volatilities
• Complexity \(L^0\)-penalized \(M\)-estimation: consistency in more dimensions
• Joint Structural Break Detection and Parameter Estimation in High-Dimensional Nonstationary VAR Models
• Optimal change-point detection and localization
• Non-smooth Variational Regularization for Processing Manifold-Valued Data
• Fused-MCP With Application to Signal Processing
• Univariate mean change point detection: penalization, CUSUM and optimality
• Multiscale change-point segmentation: beyond step functions
• A computationally efficient nonparametric approach for changepoint detection
• Locally adaptive image denoising by a statistical multiresolution criterion
• Nonparametric maximum likelihood approach to multiple change-point problems
• Convergence rates of least squares regression estimators with heavy-tailed errors
• Modified path algorithm of fused Lasso signal approximator for consistent recovery of change points
• Concentration behavior of the penalized least squares estimator
• Multiscale blind source separation
• Mumford-Shah and Potts regularization for manifold-valued data
• The $L^1$-Potts Functional for Robust Jump-Sparse Reconstruction
• Complexity penalized least squares estimators: Analytical results
• Statistical methodology in single-molecule experiments
• Jump-penalized least absolute values estimation of scalar or circle-valued signals
• Statistical challenges in microrheology
• LASSO estimation of threshold autoregressive models
• Hypothesis testing for populations of networks
• Iterative Potts minimization for the recovery of signals with discontinuities from indirect measurements: the multivariate case
• Fast and Scalable Algorithm for Detection of Structural Breaks in Big VAR Models
• On estimation of isotonic piecewise constant signals
• Exact Spike Train Inference Via $\ell_0$ Optimization
• Oracle Efficient Estimation of Structural Breaks in Cointegrating Regressions
• Wild binary segmentation for multiple change-point detection
• Two-stage data segmentation permitting multiscale change points, heavy tails and dependence
• Group LASSO for Structural Break Time Series
• Title not available (Why is that?)
• Multiscale Change Point Inference
• A Kernel Multiple Change-point Algorithm via Model Selection
• Optimal covariance change point localization in high dimensions
• FDR-control in multiscale change-point segmentation
• Robust change point detection method via adaptive LAD-Lasso
• Rank-based multiple change-point detection
• Multiple change-points detection in high dimension
• Detecting possibly frequent change-points: wild binary segmentation 2 and steepest-drop model selection
• Properties and refinements of the fused Lasso
• Changepoint detection by the quantile Lasso method
• Mining events with declassified diplomatic documents
• Estimating the Number of Block Boundaries from Diagonal Blockwise Matrices Without Penalization
• Consistency of a range of penalised cost approaches for detecting multiple changepoints
• Statistical multiresolution Dantzig estimation in imaging: fundamental concepts and algorithmic framework
• Bump detection in the presence of dependency: does it ease or does it load?
• A heuristic, iterative algorithm for change-point detection in abrupt change models
• Tail-greedy bottom-up data decompositions and fast multiple change-point detection
• Approximate \(\ell_0\)-penalized estimation of piecewise-constant signals on graphs
• Multiple change points detection in high-dimensional multivariate regression
• Detecting change points in the stress-strength reliability \(P(X < Y)\)
• Active-set based block coordinate descent algorithm in group LASSO for self-exciting threshold autoregressive model
• A sequential feature selection approach to change point detection in mean-shift change point models
• Change-point analysis for binomial autoregressive model with application to price stability counts
• Cardinality minimization, constraints, and regularization: a survey
• Multiscale Quantile Segmentation
• Threshold estimation for continuous three‐phase polynomial regression models with constant mean in the middle regime
• Scalable multiple changepoint detection for functional data sequences

Uses Software

• ftnonpar

This page was built for publication: Consistencies and rates of convergence of jump-penalized least squares estimators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1002154)