Papers
Topics
Authors
Recent
Search
2000 character limit reached

Multiscale Quantile Regression with Local Error Control

Published 17 Mar 2024 in stat.ME, math.ST, and stat.TH | (2403.11356v1)

Abstract: For robust and efficient detection of change points, we introduce a novel methodology MUSCLE (multiscale quantile segmentation controlling local error) that partitions serial data into multiple segments, each sharing a common quantile. It leverages multiple tests for quantile changes over different scales and locations, and variational estimation. Unlike the often adopted global error control, MUSCLE focuses on local errors defined on individual segments, significantly improving detection power in finding change points. Meanwhile, due to the built-in model complexity penalty, it enjoys the finite sample guarantee that its false discovery rate (or the expected proportion of falsely detected change points) is upper bounded by its unique tuning parameter. Further, we obtain the consistency and the localisation error rates in estimating change points, under mild signal-to-noise-ratio conditions. Both match (up to log factors) the minimax optimality results in the Gaussian setup. All theories hold under the only distributional assumption of serial independence. Incorporating the wavelet tree data structure, we develop an efficient dynamic programming algorithm for computing MUSCLE. Extensive simulations as well as real data applications in electrophysiology and geophysics demonstrate its competitiveness and effectiveness. An implementation via R package muscle is available from GitHub.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (65)
  1. Highly parallel computing. Benjamin-Cummings Publishing Co., Inc.
  2. Detection of an anomalous cluster in a network. Ann. Statist., 39(1):278–304.
  3. Narrowest-over-threshold detection of multiple change points and change-point-like features. J. R. Stat. Soc. Ser. B. Stat. Methodol., 81(3):649–672.
  4. Global ecosystem thresholds driven by aridity. Science, 367(6479):787–790.
  5. Sparse high-dimensional regression: exact scalable algorithms and phase transitions. Ann. Statist., 48(1):300–323.
  6. Outlier detection: Methods, models, and classification. ACM Comput. Surv., 53(3):1–37.
  7. Consistencies and rates of convergence of jump-penalized least squares estimators. Ann. Statist., 37(1):157–183.
  8. Towards optimal range medians. Theor. Comput. Sci., 412(24):2588–2601.
  9. Wavelet trees for competitive programming. Olymp. Inform., 10:19–37.
  10. Introduction to algorithms. MIT Press, Cambridge, MA, third edition.
  11. Inferring change points in the spread of covid-19 reveals the effectiveness of interventions. Science, 369(6500):eabb9789.
  12. Multiscale change point detection for dependent data. Scand. J. Stat., 47(4):1243–1274.
  13. Ideal spatial adaptation by wavelet shrinkage. Biometrika, 81(3):425–455.
  14. Stepwise signal extraction via marginal likelihood. J. Amer. Statist. Assoc., 111(513):314–330.
  15. On the limit distribution of multiscale test statistics for nonparametric curve estimation. Math. Methods Statist., 15(1):20–25.
  16. Multiscale testing of qualitative hypotheses. Ann. Statist., 29(1):124–152.
  17. A MOSUM procedure for the estimation of multiple random change points. Bernoulli, 24(1):526–564.
  18. Window size selection in unsupervised time series analytics: A review and benchmark. In Guyet, T., Ifrim, G., Malinowski, S., Bagnall, A., Shafer, P., and Lemaire, V., editors, Advanced Analytics and Learning on Temporal Data, pages 83–101, Cham. Springer International Publishing.
  19. Identifying localized changes in large systems: Change-point detection for biomolecular simulations. Proc. Natl. Acad. Sci. U.S.A., 112(24):7454–7459.
  20. Segmentation and estimation of change-point models: false positive control and confidence regions. Ann. Statist., 48(3):1615–1647.
  21. Fearnhead, P. (2006). Exact and efficient Bayesian inference for multiple changepoint problems. Stat. Comput., 16(2):203–213.
  22. Changepoint detection in the presence of outliers. J. Amer. Statist. Assoc., 114(525):169–183.
  23. Multiscale change point inference. J. R. Stat. Soc. Ser. B. Stat. Methodol., 76(3):495–580.
  24. Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection. Ann. Statist., 42(6):2243–2281.
  25. Fryzlewicz, P. (2024). Robust narrowest significance pursuit: Inference for multiple change-points in the median. J. Bus. Econ. Stat. To appear.
  26. Multiscale DNA partitioning: statistical evidence for segments. Bioinformatics, 30(16):2255–2262.
  27. Mathematical foundations of infinite-dimensional statistical models. Cambridge University Press, New York.
  28. High-order entropy-compressed text indexes. In Proceedings of the Fourteenth Annual ACM-SIAM Symposium on Discrete Algorithms, page 841–850. Society for Industrial and Applied Mathematics.
  29. A variational view on statistical multiscale estimation. Annu. Rev. Stat. Appl., 9:343–372.
  30. Multiple change-point estimation with a total variation penalty. J. Amer. Statist. Assoc., 105(492):1480–1493.
  31. A computationally efficient nonparametric approach for changepoint detection. Stat. Comput., 27(5):1293–1305.
  32. Monitoring for a change point in a sequence of distributions. Ann. Statist., 49(4):2271–2291.
  33. Kabluchko, Z. (2011). Extremes of the standardized Gaussian noise. Stochastic Process. Appl., 121(3):515–533.
  34. Optimal detection of changepoints with a linear computational cost. J. Amer. Statist. Assoc., 107(500):1590–1598.
  35. Koenker, R. (2005). Quantile Regression. Cambridge University Press.
  36. Seeded binary segmentation: a general methodology for fast and optimal changepoint detection. Biometrika, 110(1):249–256.
  37. Multiscale change-point segmentation: beyond step functions. Electron. J. Stat., 13(2):3254–3296.
  38. FDR-control in multiscale change-point segmentation. Electron. J. Stat., 10(1):918–959.
  39. Change-point detection in time-series data by relative density-ratio estimation. Neural Netw., 43:72–83.
  40. Optimal nonparametric change point analysis. Electron. J. Stat., 15(1):1154–1201.
  41. On optimal multiple changepoint algorithms for large data. Stat. Comput., 27(2):519–533.
  42. A nonparametric approach for multiple change point analysis of multivariate data. J. Amer. Statist. Assoc., 109(505):334–345.
  43. Navarro, G. (2014). Wavelet trees for all. J. Discrete Algorithms, 25:2–20.
  44. Multiple change-point detection: a selective overview. Statist. Sci., 31(4):611–623.
  45. Norris, J. R. (1998). Markov chains, volume 2. Cambridge University Press, Cambridge.
  46. Page, E. S. (1955). A test for a change in a parameter occurring at an unknown point. Biometrika, 42:523–527.
  47. Heterogeneous idealization of ion channel recordings–open channel noise. IEEE Trans. Nanobioscience, 20(1):57–78.
  48. Heterogeneous change point inference. J. R. Stat. Soc. Ser. B. Stat. Methodol., 79(4):1207–1227.
  49. Fully automatic multiresolution idealization for filtered ion channel recordings: flickering event detection. IEEE Trans. Nanobioscience, 17(3):300–320.
  50. V-measure: A conditional entropy-based external cluster evaluation measure. In Proceedings of the 2007 joint conference on empirical methods in natural language processing and computational natural language learning (EMNLP-CoNLL), pages 410–420.
  51. Numerical Bayesian methods applied to signal processing. Springer Science & Business Media.
  52. Breaks and the statistical process of inflation: the case of estimating the ‘modern’long-run phillips curve. Empir. Econ., 56:1455–1475.
  53. Sakmann, B. (2013). Single-channel recording. Springer Science & Business Media.
  54. Selective review of offline change point detection methods. Signal Process., 167:107299.
  55. Tsybakov, A. B. (2009). Introduction to nonparametric estimation. Springer Series in Statistics. Springer, New York.
  56. Multiscale quantile segmentation. J. Amer. Statist. Assoc., 117(539):1384–1397.
  57. Identification of hidden markov models for ion channel currents. II. State-dependent excess noise. IEEE Trans. Signal Process., 46(7):1916–1929.
  58. Optimal change-point detection and localization. Ann. Statist., 51(4):1586–1610.
  59. Vostrikova, L. Y. (1981). Detecting “disorder” in multidimensional random processes. In Doklady akademii nauk, volume 259, pages 270–274. Russian Academy of Sciences.
  60. Wald, A. (1945). Sequential tests of statistical hypotheses. Ann. Math. Statist., 16:117–186.
  61. Calibrating the scan statistic: finite sample performance versus asymptotics. J. R. Stat. Soc. Ser. B. Stat. Methodol., 84(5):1608–1639.
  62. Wu, W. B. (2005). Nonlinear system theory: another look at dependence. Proc. Natl. Acad. Sci. USA, 102(40):14150–14154.
  63. Approximate simulation-free Bayesian inference for multiple changepoint models with dependence within segments. Bayesian Anal., 6(4):501–527.
  64. Nonparametric maximum likelihood approach to multiple change-point problems. Ann. Statist., 42(3):970–1002.
  65. Composite quantile regression and the oracle model selection theory. Ann. Statist., 36(3):1108–1126.
Citations (1)
View on Semantic Scholar

Summary

No one has generated a summary of this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Sign Up to Summarize

Paper to Video (Beta)

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now

Authors (2)

  1. Zhi Liu 
  2. Housen Li 

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up

Tweets

Sign up for free to view the 1 tweet with 0 likes about this paper.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up for Free