The Annals of Statistics

On optimality of the Shiryaev–Roberts procedure for detecting a change in distribution

Aleksey S. Polunchenko and Alexander G. Tartakovsky

Full-text: Open access

Abstract

In 1985, for detecting a change in distribution, Pollak introduced a specific minimax performance metric and a randomized version of the Shiryaev–Roberts procedure where the zero initial condition is replaced by a random variable sampled from the quasi-stationary distribution of the Shiryaev–Roberts statistic. Pollak proved that this procedure is third-order asymptotically optimal as the mean time to false alarm becomes large. The question of whether Pollak’s procedure is strictly minimax for any false alarm rate has been open for more than two decades, and there were several attempts to prove this strict optimality. In this paper, we provide a counterexample which shows that Pollak’s procedure is not optimal and that there is a strictly optimal procedure which is nothing but the Shiryaev–Roberts procedure that starts with a specially designed deterministic point.

Article information

Source
Ann. Statist., Volume 38, Number 6 (2010), 3445-3457.

Dates
First available in Project Euclid: 30 November 2010

Permanent link to this document
https://projecteuclid.org/euclid.aos/1291126963

Digital Object Identifier
doi:10.1214/09-AOS775

Mathematical Reviews number (MathSciNet)
MR2766858

Zentralblatt MATH identifier
1204.62141

Subjects
Primary: 62L10: Sequential analysis 62L15: Optimal stopping [See also 60G40, 91A60]
Secondary: 60G40: Stopping times; optimal stopping problems; gambling theory [See also 62L15, 91A60]

Keywords
Changepoint problems Shiryaev–Roberts procedures sequential detection

Citation

Polunchenko, Aleksey S.; Tartakovsky, Alexander G. On optimality of the Shiryaev–Roberts procedure for detecting a change in distribution. Ann. Statist. 38 (2010), no. 6, 3445--3457. doi:10.1214/09-AOS775. https://projecteuclid.org/euclid.aos/1291126963


Export citation

References

  • Feinberg, E. A. and Shiryaev, A. N. (2006). Quickest detection of drift change for Brownian motion in generalized Bayesian and minimax settings. Statist. Decisions 24 445–470.
  • Harris, T. E. (1963). The Theory of Branching Processes. Springer, Berlin.
  • Mei, Y. (2006). Comments on, “A note on optimal detection of a change in distribution,” by Benjamin Yakir. Ann. Statist. 34 1570–1576.
  • Moustakides, G. V., Polunchenko, A. S. and Tartakovsky, A. G. (2009). Numerical comparison of CUSUM and Shiryaev–Roberts procedures for detecting changes in distributions. Comm. Statist. Theory Methods 38 3225–3239.
  • Moustakides, G. V., Polunchenko, A. S. and Tartakovsky, A. G. (2011). A numerical approach to comparative efficiency analysis of quickest change-point detection procedures. Statist. Sinica. 21. To appear.
  • Pollak, M. (1985). Optimal detection of a change in distribution. Ann. Statist. 13 206–227.
  • Pollak, M. (1987). Average run lengths of an optimal method of detecting a change in distribution. Ann. Statist. 15 749–779.
  • Pollak, M. and Tartakovsky, A. G. (2009). Optimality properties of the Shiryaev–Roberts procedure. Statist. Sinica 19 1729–1739.
  • Roberts, S. W. (1966). A comparison of some control chart procedures. Technometrics 8 411–430.
  • Shiryaev, A. N. (1961). The problem of the most rapid detection of a disturbance in a stationary process. Dokl. Math. 2 795–799.
  • Shiryaev, A. N. (1963). On optimum methods in quickest detection problems. Theory Probab. Appl. 8 22–46.
  • Yakir, B. (1997). A note on optimal detection of a change in distribution. Ann. Statist. 25 2117–2126.