The Annals of Statistics

Super-resolution estimation of cyclic arrival rates

Ningyuan Chen, Donald K. K. Lee, and Sahand N. Negahban

Full-text: Open access

Abstract

Exploiting the fact that most arrival processes exhibit cyclic behaviour, we propose a simple procedure for estimating the intensity of a nonhomogeneous Poisson process. The estimator is the super-resolution analogue to Shao (2010) and Shao and Lii [J. R. Stat. Soc. Ser. B. Stat. Methodol. 73 (2011) 99–122], which is a sum of $p$ sinusoids where $p$ and the amplitude and phase of each wave are not known and need to be estimated. This results in an interpretable yet flexible specification that is suitable for use in modelling as well as in high resolution simulations.

Our estimation procedure sits in between classic periodogram methods and atomic/total variation norm thresholding. Through a novel use of window functions in the point process domain, our approach attains super-resolution without semidefinite programming. Under suitable conditions, finite sample guarantees can be derived for our procedure. These resolve some open questions and expand existing results in spectral estimation literature.

Article information

Source
Ann. Statist., Volume 47, Number 3 (2019), 1754-1775.

Dates
Received: June 2017
Revised: June 2018
First available in Project Euclid: 13 February 2019

Permanent link to this document
https://projecteuclid.org/euclid.aos/1550026856

Digital Object Identifier
doi:10.1214/18-AOS1736

Mathematical Reviews number (MathSciNet)
MR3911129

Zentralblatt MATH identifier
07053525

Subjects
Primary: 62M15: Spectral analysis 90B22: Queues and service [See also 60K25, 68M20]
Secondary: 60G55: Point processes

Keywords
Spectral estimation periodogram window function thresholding nonhomogeneous Poisson process queueing theory

Citation

Chen, Ningyuan; Lee, Donald K. K.; Negahban, Sahand N. Super-resolution estimation of cyclic arrival rates. Ann. Statist. 47 (2019), no. 3, 1754--1775. doi:10.1214/18-AOS1736. https://projecteuclid.org/euclid.aos/1550026856


Export citation

References

  • Bartlett, M. S. (1963). The spectral analysis of point processes. J. Roy. Statist. Soc. Ser. B 25 264–296.
  • Bebbington, M. and Zitikis, R. (2004). A robust heuristic estimator for the period of a Poisson intensity function. Methodol. Comput. Appl. Probab. 6 441–462.
  • Belitser, E., Serra, P. and van Zanten, H. (2013). Estimating the period of a cyclic non-homogeneous Poisson process. Scand. J. Stat. 40 204–218.
  • Bhaskar, B. N., Tang, G. and Recht, B. (2013). Atomic norm denoising with applications to line spectral estimation. IEEE Trans. Signal Process. 61 5987–5999.
  • Brown, L., Cai, T., Zhang, R., Zhao, L. and Zhou, H. (2010). The root-unroot algorithm for density estimation as implemented via wavelet block thresholding. Probab. Theory Related Fields 146 401–433.
  • Candès, E. J. and Fernandez-Granda, C. (2013). Super-resolution from noisy data. J. Fourier Anal. Appl. 19 1229–1254.
  • Chen, N., Lee, D. K. K. and Negahban, S. N. (2019). Supplement to “Super-resolution estimation of cyclic arrival rates.” DOI:10.1214/18-AOS1736SUPP.
  • Chen, N., Lee, D. K. K. and Shen, H. (2018). Can customer arrival rates be modelled by sine waves? Working paper.
  • Donoho, D. L. and Johnstone, I. M. (1994). Ideal spatial adaptation by wavelet shrinkage. Biometrika 81 425–455.
  • Dutt, A. and Rokhlin, V. (1993). Fast Fourier transforms for nonequispaced data. SIAM J. Sci. Comput. 14 1368–1393.
  • Fernandez-Granda, C. (2013). Support detection in super-resolution. In Proceedings of the 10th International Conference on Sampling Theory and Applications 145–148.
  • Helmers, R. and Mangku, I. W. (2003). On estimating the period of a cyclic Poisson process. In Mathematical Statistics and Applications: Festschrift for Constance van Eeden. Institute of Mathematical Statistics Lecture Notes—Monograph Series 42 345–356. IMS, Beachwood, OH.
  • Lewis, P. (1970). Remarks on the theory, computation and application of the spectral analysis of series of events. J. Sound Vibration 12 353–375.
  • Li, T.-H. (2014). Time Series with Mixed Spectra. CRC Press, Boca Raton, FL.
  • Moitra, A. (2015). Super-resolution, extremal functions and the condition number of Vandermonde matrices. In STOC’15—Proceedings of the 2015 ACM Symposium on Theory of Computing 821–830. ACM, New York.
  • Osipov, A., Rokhlin, V. and Xiao, H. (2013). Prolate Spheroidal Wave Functions of Order Zero: Mathematical Tools for Bandlimited Approximation. Applied Mathematical Sciences 187. Springer, New York.
  • Prabhu, K. (2013). Window Functions and Their Applications in Signal Processing. CRC Press, New York.
  • Rice, J. A. and Rosenblatt, M. (1988). On frequency estimation. Biometrika 75 477–484.
  • Shao, N. (2010). Modeling almost periodicity in point processes. Ph.D. thesis, Univ. California, Riverside.
  • Shao, N. and Lii, K.-S. (2011). Modelling non-homogeneous Poisson processes with almost periodic intensity functions. J. R. Stat. Soc. Ser. B. Stat. Methodol. 73 99–122.
  • Tang, G., Bhaskar, B. N. and Recht, B. (2015). Near minimax line spectral estimation. IEEE Trans. Inform. Theory 61 499–512.
  • Vere-Jones, D. (1982). On the estimation of frequency in point-process data. J. Appl. Probab. 19A 383–394.

Supplemental materials

  • Proofs and asymptotic normality. The proofs of all results presented in this paper are provided in Appendix A of the supplement. Appendix B establishes the asymptotic normality of the windowed periodogram estimator.