Approximate Recovery in Changepoint Problems, from $\ell_2$ Estimation Error Rates

In the 1-dimensional multiple changepoint detection problem, we prove that any procedure with a fast enough $\ell_2$ error rate, in terms of its estimation of the underlying piecewise constant mean vector, automatically has an (approximate) changepoint screening property---specifically, each true jump in the underlying mean vector has an estimated jump nearby. We also show, again assuming only knowledge of the $\ell_2$ error rate, that a simple post-processing step can be used to eliminate spurious estimated changepoints, and thus delivers an (approximate) changepoint recovery property---specifically, in addition to the screening property described above, we are assured that each estimated jump has a true jump nearby. As a special case, we focus on the application of these results to the 1-dimensional fused lasso, i.e., 1-dimensional total variation denoising, and compare the implications with existing results from the literature. We also study extensions to related problems, such as changepoint detection over graphs.

[1]  S. Geer.  Estimating a Regression Function , 1990 .

[2]  L. Rudin,et al.  Nonlinear total variation based noise removal algorithms , 1992 .

[3]  I. Johnstone,et al.  Ideal spatial adaptation by wavelet shrinkage , 1994 .

[4]  S. Geer,et al.  Locally adaptive regression splines , 1997 .

[5]  I. Johnstone,et al.  Minimax estimation via wavelet shrinkage , 1998 .

[6]  P. Davies,et al.  Local Extremes, Runs, Strings and Multiresolution , 2001 .

[7]  S. R. Jammalamadaka,et al.  Empirical Processes in M-Estimation , 2001 .

[8]  M. Talagrand.  The Generic chaining : upper and lower bounds of stochastic processes , 2005 .

[9]  R. Tibshirani,et al.  Sparsity and smoothness via the fused lasso , 2005 .

[10]  Stephan Didas,et al.  Splines in Higher Order TV Regularization , 2006, International Journal of Computer Vision.

[11]  P. Fryzlewicz.  Unbalanced Haar Technique for Nonparametric Function Estimation , 2007 .

[12]  A. Rinaldo.  Properties and refinements of the fused lasso , 2008, 0805.0234.

[13]  L. Duembgen,et al.  Multiscale inference about a density , 2007, 0706.3968.

[14]  Stephen P. Boyd,et al.  1 Trend Filtering , 2009, SIAM Rev..

[15]  Holger Hoefling.  A Path Algorithm for the Fused Lasso Signal Approximator , 2009, 0910.0526.

[16]  V. Liebscher,et al.  Consistencies and rates of convergence of jump-penalized least squares estimators , 2009, 0902.4838.

[17]  Z. Harchaoui,et al.  Multiple Change-Point Estimation With a Total Variation Penalty , 2010 .

[18]  H. Chan,et al.  Detection with the scan and the average likelihood ratio , 2011, 1107.4344.

[19]  R. Tibshirani,et al.  The solution path of the generalized lasso , 2010, 1005.1971.

[20]  Junyang Qian,et al.  On pattern recovery of the fused Lasso , 2012, 1211.5194.

[21]  Alessandro Rinaldo,et al.  Sparsistency of the Edge Lasso over Graphs , 2012, AISTATS.

[22]  A. Munk,et al.  Multiscale change point inference , 2013, 1301.7212.

[23]  R. Tibshirani.  Adaptive piecewise polynomial estimation via trend filtering , 2013, 1304.2986.

[24]  Piotr Fryzlewicz,et al.  Wild binary segmentation for multiple change-point detection , 2014, 1411.0858.

[25]  B. Wahlberg,et al.  Submitted to the Annals of Statistics ON CHANGE POINT DETECTION USING THE FUSED LASSO METHOD ∗ By , 2014 .

[26]  A. Dalalyan,et al.  On the Prediction Performance of the Lasso , 2014, 1402.1700.

[27]  Jan-Christian Hü,et al.  Optimal rates for total variation denoising , 2016, COLT.

[28]  Yu-Xiang Wang,et al.  Total Variation Classes Beyond 1d: Minimax Rates, and the Limitations of Linear Smoothers , 2016, NIPS.

[29]  Alexander J. Smola,et al.  Trend Filtering on Graphs , 2014, J. Mach. Learn. Res..

[30]  P. Fryzlewicz.  Tail-greedy bottom-up data decompositions and fast multiple change-point detection , 2018, The Annals of Statistics.