The Annals of Statistics

Change-point in stochastic design regression and the bootstrap

Emilio Seijo and Bodhisattva Sen

Full-text: Open access

Abstract

In this paper we study the consistency of different bootstrap procedures for constructing confidence intervals (CIs) for the unique jump discontinuity (change-point) in an otherwise smooth regression function in a stochastic design setting. This problem exhibits nonstandard asymptotics, and we argue that the standard bootstrap procedures in regression fail to provide valid confidence intervals for the change-point. We propose a version of smoothed bootstrap, illustrate its remarkable finite sample performance in our simulation study and prove the consistency of the procedure. The m out of n bootstrap procedure is also considered and shown to be consistent. We also provide sufficient conditions for any bootstrap procedure to be consistent in this scenario.

Article information

Source
Ann. Statist., Volume 39, Number 3 (2011), 1580-1607.

Dates
First available in Project Euclid: 7 June 2011

Permanent link to this document
https://projecteuclid.org/euclid.aos/1307452129

Digital Object Identifier
doi:10.1214/11-AOS874

Mathematical Reviews number (MathSciNet)
MR2850213

Zentralblatt MATH identifier
1220.62043

Subjects
Primary: 62G08: Nonparametric regression 62G05: Estimation
Secondary: 62G20: Asymptotic properties

Keywords
Argmax continuous mapping theorem consistency of the bootstrap m out of n bootstrap nonstandard asymptotics semiparametric regression smoothed bootstrap

Citation

Seijo, Emilio; Sen, Bodhisattva. Change-point in stochastic design regression and the bootstrap. Ann. Statist. 39 (2011), no. 3, 1580--1607. doi:10.1214/11-AOS874. https://projecteuclid.org/euclid.aos/1307452129


Export citation

References

  • [1] Abrevaya, J. and Huang, J. (2005). On the bootstrap of the maximum score estimator. Econometrica 73 1175–1204.
  • [2] Bickel, P. J., Götze, F. and van Zwet, W. R. (1997). Resampling fewer than n observations: Gains, losses, and remedies for losses. Statist. Sinica 7 1–31.
  • [3] Billingsley, P. (1986). Probability and Measure. Wiley, New York.
  • [4] Bose, A. and Chatterjee, S. (2001). Generalised bootstrap in non-regular M-estimation problems. Statist. Probab. Lett. 55 319–328.
  • [5] Dempfle, A. and Stute, W. (2002). Nonparametric estimation of a discontinuity in regression. Statist. Neerlandica 56 233–242.
  • [6] Dudley, R. M. (2002). Real Analysis and Probability. Cambridge Univ. Press, Cambridge.
  • [7] Dümbgen, L. (1991). The asymptotic behavior of some nonparametric change-point estimators. Ann. Statist. 19 1471–1495.
  • [8] Efron, B. (1982). The Jackknife, the Bootstrap and Other Resampling Plans. CBMS-NSF Regional Conference Series in Applied Mathematics 38. SIAM, Philadelphia, PA.
  • [9] Ethier, S. and Kurtz, T. (2005). Markov Processes, Characterization and Convergence. Wiley, New York.
  • [10] Freedman, D. A. (1981). Bootstrapping regression models. Ann. Statist. 9 1218–1228.
  • [11] Gijbels, I., Hall, P. and Kneip, A. (1999). On the estimation of jump point in smooth curves. Ann. Inst. Statist. Math. 51 231–251.
  • [12] Gijbels, I., Hall, P. and Kneip, A. (2004). Interval and band estimation for curves with jumps. J. Appl. Probab. 41 65–79.
  • [13] Hǔsková, M. and Kirch, C. (2008). Bootstrapping confidence intervals for the change-point of time series. J. Time Ser. Anal. 29 947–972.
  • [14] Kosorok, M. R. (2008). Bootstrapping in Grenander estimator. In Beyond Parametrics in Interdisciplinary Research: Festschrift in Honor of Professor Pranab K. Sen. Inst. Math. Stat. Collect. 1 282–292. IMS, Beachwood, OH.
  • [15] Kosorok, M. R. (2008). Introduction to Empirical Processes and Semiparametric Inference. Springer, New York.
  • [16] Kosorok, M. R. and Song, R. (2007). Inference under right censoring for transformation models with a change-point based on a covariate threshold. Ann. Statist. 35 957–989.
  • [17] Koul, H., Qian, L. and Surgailis, D. (2003). Asymptotics of M-estimators in two-phase linear regression models. Stochastic Process. Appl. 103 123–154.
  • [18] Lan, Y., Banerjee, M. and Michailidis, G. (2009). Change-point estimation under adaptive sampling. Ann. Statist. 37 1752–1791.
  • [19] Lee, S. M. S. and Pun, M. C. (2006). On m out of n bootstrapping for nonstandard M-estimation with nuisance parameters. J. Amer. Statist. Assoc. 101 1185–1197.
  • [20] Möhle, M. (2005). Convergence results for compound Poisson distributions and applications to the standard Luria–Delbrück distribution. J. Appl. Probab. 42 620–631.
  • [21] Neuhaus, G. (1971). On weak convergence of stochastic processes with multidimensional time parameter. Ann. Math. Statist. 42 1285–1295.
  • [22] Politis, D., Romano, J. and Wolf, W. (1999). Subsampling. Springer, New York.
  • [23] Pons, O. (2003). Estimation in a Cox regression model with a change-point according to a threshold covariate. Ann. Statist. 31 442–463.
  • [24] Scott, D. W. (1992). Multivariate Density Estimation: Theory, Practice, and Visualization. Wiley, New York.
  • [25] Seijo, E. and Sen, B. (2011). Supplement to “Change-point in stochastic design regression and the bootstrap.” DOI:10.1214/11-AOS874SUPP.
  • [26] Sen, B., Banerjee, M. and Woodroofe, M. (2010). Inconsistency of bootstrap: The Grenander estimator. Ann. Statist. 38 1953–1977.
  • [27] Shao, J. and Tu, D. (1995). The Jackknife and Bootstrap. Springer, New York.
  • [28] Silverman, B. W. (1978). Weak and strong uniform consistency of the kernel estimate of a density and its derivatives. Ann. Statist. 6 177–184.
  • [29] Wu, C. F. J. (1986). Jackknife, bootstrap and other resampling methods in regression analysis. Ann. Statist. 14 1261–1295.

Supplemental materials

  • Supplementary material: Supplement to “Change-point in stochastic design regression and the bootstrap”. The supplementary file contains a longer version of this paper with all the technical details which were excluded in the present version due to their length.