An Accurate Method for Determining the Pre-Change Run Length Distribution of the Generalized Shiryaev-Roberts Detection Procedure

Abstract Change-of-measure is a powerful technique in wide use across statistics, probability, and analysis. Particularly known as Wald's likelihood ratio identity, the technique enabled the proof of a number of exact and asymptotic optimality results pertaining to the problem of quickest change-point detection. Within the latter problem's context we apply the technique to develop a numerical method to compute the generalized Shiryaev–Roberts (GSR) detection procedure's pre-change run length distribution. Specifically, the method is based on the integral equations approach and uses the collocation framework with the basis functions chosen to exploit a certain change-of-measure identity and a specific martingale property of the GSR procedure's detection statistic. As a result, the method's accuracy and robustness improve substantially, even though the method's theoretical rate of convergence is shown to be merely quadratic. A tight upper bound on the method's error is supplied as well. The method is not restricted to a particular data distribution or to a specific value of the GSR detection statistic's head start. To conclude, we offer a case study to demonstrate the proposed method at work, drawing particular attention to the method's accuracy and its robustness with respect to three factors: (1) partition size (rough vs. fine), (2) change magnitude (faint vs. contrast), and (3) average run length (ARL) to false alarm level (low vs. high). Specifically, assuming independent standard Gaussian observations undergoing a surge in the mean, we employ the method to study the GSR procedure's run length's pre-change distribution, its average (i.e., the usual ARL to false alarm), and its standard deviation. As expected from the theoretical analysis, the method's high accuracy and robustness with respect to the foregoing three factors are confirmed experimentally. We also comment on extending the method to handle other performance measures and other procedures.

[1]  William H. Woodall,et al.  The Distribution of the Run Length of One-Sided CUSUM Procedures for Continuous Random Variables , 1983 .

[2]  Larry Goldstein,et al.  Size bias, sampling, the waiting time paradox, and infinite divisibility: when is the increment independent? , 2010, 1007.3910.

[3]  E. S. Page.  CONTINUOUS INSPECTION SCHEMES , 1954 .

[4]  K. Fan.  Topological proofs for certain theorems on matrices with non-negative elements , 1958 .

[5]  M. Pollak.  Average Run Lengths of an Optimal Method of Detecting a Change in Distribution. , 1987 .

[6]  A. Tartakovsky.  Asymptotic Performance of a Multichart CUSUM Test Under False Alarm Probability Constraint , 2005, Proceedings of the 44th IEEE Conference on Decision and Control.

[7]  Moshe Pollak,et al.  Third-order Asymptotic Optimality of the Generalized Shiryaev-Roberts Changepoint Detection Procedures , 2010, 1005.1129.

[8]  G. Moustakides,et al.  State-of-the-Art in Bayesian Changepoint Detection , 2010 .

[9]  B. Yakir.  Discussion on “Is Average Run Length to False Alarm Always an Informative Criterion?” by Yajun Mei , 2008 .

[10]  M. A. Girshick,et al.  A BAYES APPROACH TO A QUALITY CONTROL MODEL , 1952 .

[11]  Ronald B. Crosier,et al.  Fast Initial Response for CUSUM Quality-Control Schemes: Give Your CUSUM A Head Start.: Give Your CUSUM A Head Start. , 2000 .

[12]  Thomas P. Ryan,et al.  Statistical methods for quality improvement , 1989 .

[13]  Stephen A. Freitas,et al.  Modern Industrial Statistics: Design and Control of Quality and Reliability , 1999, Technometrics.

[14]  A. Shiryaev.  On Optimum Methods in Quickest Detection Problems , 1963 .

[15]  Kluwer Academic Publishers.  Methodology and computing in applied probability , 1999 .

[16]  S. W. Roberts,et al.  Control Chart Tests Based on Geometric Moving Averages , 2000, Technometrics.

[17]  Design and Comparison of Shiryaev-Roberts and CUSUM-Type Change-Point Detection Procedures , 2009 .

[18]  Douglas C. Montgomery,et al.  Introduction to Statistical Quality Control , 1986 .

[19]  D. Siegmund.  Sequential Analysis: Tests and Confidence Intervals , 1985 .

[20]  A. Tartakovsky,et al.  A Nonparametric Multichart CUSUM Test for Rapid Intrusion Detection , 2007 .

[21]  Moshe Pollak,et al.  A Small Sample Size Comparison of the Cusum and Shiryayev-Roberts Approaches: Changepoint Detection , 1991 .

[22]  E. Yashchin.  Statistical Control Schemes: Methods, Applications and Generalizations , 1993 .

[23]  George V. Moustakides,et al.  Numerical Comparison of CUSUM and Shiryaev–Roberts Procedures for Detecting Changes in Distributions , 2009, 0908.4119.

[24]  William H. Woodall,et al.  Performance comparison of some likelihood ratio-based statistical surveillance methods , 2008 .

[25]  D. A. Evans,et al.  An approach to the probability distribution of cusum run length , 1972 .

[26]  Alexander G. Tartakovsky,et al.  Nearly Optimal Change-Point Detection with an Application to Cybersecurity , 2012, 1202.2849.

[27]  T. Lai.  Likelihood Ratio Identities and Their Applications to Sequential Analysis , 2004 .

[28]  Moshe Pollak,et al.  Asymptotic Exponentiality of the Distribution of First Exit Times for a Class of Markov Processes with Applications to Quickest Change Detection , 2006 .

[29]  Tze Leung Lai,et al.  Information Bounds and Quick Detection of Parameter Changes in Stochastic Systems , 1998, IEEE Trans. Inf. Theory.

[30]  Michèle Basseville,et al.  Detection of abrupt changes: theory and application , 1993 .

[31]  Thomas P. Ryan,et al.  Statistical Methods for Quality Improvement: Ryan/Quality Improvement 3E , 2011 .

[32]  S. W. Roberts.  A Comparison of Some Control Chart Procedures , 1966 .

[33]  Michael Woodroofe,et al.  Nonlinear Renewal Theory , 2006 .

[34]  Charles W. Champ,et al.  A a comparison of the markov chain and the integral equation approaches for evaluating the run length distribution of quality control charts , 1991 .

[35]  M. Woodroofe.  Nonlinear Renewal Theory in Sequential Analysis , 1987 .

[36]  P. Rabinowitz.  Russian Numerical Analysis: Approximate Methods of Higher Analysis . L. V. Kantorovich and V. I. Krylov. Translated from the third Russian edition by Curtis D. Benster. Interscience, New York, 1959. xv + 681. $17. , 1961, Science.

[37]  S. Banach.  Sur les opérations dans les ensembles abstraits et leur application aux équations intégrales , 1922 .

[38]  Asymptotic performance of a multistate coherent system , 1988, Advances in Applied Probability.

[39]  Rudolf B. Blazek,et al.  Detection of intrusions in information systems by sequential change-point methods , 2005 .

[40]  K. Atkinson,et al.  Theoretical Numerical Analysis: A Functional Analysis Framework , 2001 .

[41]  N. Kryloff,et al.  La solution approchée du problème de Dirichlet , 1930 .

[42]  M. Pollak.  Optimal Detection of a Change in Distribution , 1985 .

[43]  G. Lorden.  PROCEDURES FOR REACTING TO A CHANGE IN DISTRIBUTION , 1971 .

[44]  Alexander G. Tartakovsky.  Discussion on “Is Average Run Length to False Alarm Always an Informative Criterion?” by Yajun Mei , 2008 .

[45]  Aleksey S. Polunchenko,et al.  State-of-the-Art in Sequential Change-Point Detection , 2011, 1109.2938.

[46]  Grigory Sokolov,et al.  Quickest Change-Point Detection: A Bird's Eye View , 2013, 1310.3285.

[47]  Moshe Pollak,et al.  Asymptotic Exponentiality of First Exit Times for Recurrent Markov Processes and Applications to Changepoint Detection , 2008 .

[48]  Alexander G. Tartakovsky,et al.  On optimality of the Shiryaev-Roberts procedure for detecting a change in distribution , 2009, 0904.3370.

[49]  Sven Knoth,et al.  The Art of Evaluating Monitoring Schemes — How to Measure the Performance of Control Charts? , 2006 .

[50]  Moshe Pollak,et al.  ON OPTIMALITY PROPERTIES OF THE SHIRYAEV-ROBERTS PROCEDURE , 2007, 0710.5935.

[51]  D. Siegmund,et al.  APPROXIMATIONS TO THE EXPECTED SAMPLE SIZE OF CERTAIN SEQUENTIAL TESTS , 1975 .

[52]  James M. Lucas,et al.  Fast Initial Response for CUSUM Quality-Control Schemes: Give Your CUSUM A Head Start , 2000, Technometrics.

[53]  Alexander G. Tartakovsky,et al.  Efficient Computer Network Anomaly Detection by Changepoint Detection Methods , 2012, IEEE Journal of Selected Topics in Signal Processing.

[54]  George V. Moustakides,et al.  A NUMERICAL APPROACH TO PERFORMANCE ANALYSIS OF QUICKEST CHANGE-POINT DETECTION PROCEDURES , 2009, 0907.3521.