Abstract
In this article we investigate the performance of robust scan statistics based on moving medians, as test statistics for detecting a local change in population mean, for one and two dimensional data. When a local change in the population mean has not occurred and outliers are not present in the data, we derive approximations for the tail probabilities of fixed window scan statistics based on moving medians. The performance of the proposed robust scan statistics are evaluated and compared to, via power calculations, the performance of scan statistics based on moving sums, that have been previously investigated in the statistical literature. Numerical results based on a simulation study, for independent and identically distributed (iid) normal observations with known variance, indicate that in presence of outliers the scan statistic based on moving medians outperform the scan statistic based on moving sums, in terms of achieving more accurately the specified probability of type I error. The performance of a multiple window scan statistic based on moving medians for detecting a local change in population mean, for one and two dimensional normal data in presence of outliers, when the size of the window where a change has occurred is unknown has been investigated as well.
Similar content being viewed by others
References
Alm SE (1999) Approximations of the distributions of scan statistics of Poisson processes. In: Scan statistics and applications. Springer, New York, pp 113–139
Balakrishnan N, Koutras MV (2011) Runs and scans with applications, vol 764. Wiley, New York
Bauer P, Hackl P (1978) The use of MOSUMS for quality control. Technometrics 20(4):431–436
Bauer P, Hackl P (1980) An extension of the MOSUM technique for quality control. Technometrics 22(1):1–7
Boutsikas MV, Koutras MV (2000) Reliability approximation for Markov chain imbeddable systems. Methodol Comput Appl Probab 2(4):393–411
Buzzi-Ferraris G, Manenti F (2011) Outlier detection in large data sets. Comput Chem Eng 35(2):388–390
Cressie N (2015) Statistics for spatial data. Wiley, New York
Darling R, Waterman MS (1986) Extreme value distribution for the largest cube in a random lattice. SIAM J Appl Math 46(1):118–132
Do Lago CL, Juliano VF, Kascheres C (1995) Applying moving median digital filter to mass spectrometry and potentiometric titration. Anal Chim Acta 310 (2):281–288
Fu J, Koutras M (1994) Distribution theory of runs: a Markov chain approach. J Am Stat Assoc 89(427):1050–1058
Fu JC, Lou WW (2003) Distribution theory of runs and patterns and its applications: a finite markov chain imbedding approach. World Scientific, Singapore
Glaz J, Balakrishnan N (1999) Introduction to scan statistics. In: Scan statistics and applications. Springer, New York, pp 3–24
Glaz J, Johnson B (1988) Boundary crossing for moving sums. J Appl Probab 25(1):81–88
Glaz J, Naus J (1991) Tight bounds and approximations for scan statistic probabilities for discrete data. Ann Appl Probab 1(2):306–318
Glaz J, Naus JI, Wallenstein S (2001) Scan statistics. Springer, New York
Glaz J, Pozdnyakov V, Wallenstein S (2009) Scan statistics: methods and applications. Springer Science & Business Media, Berlin
Glaz J, Naus J, Wang X (2012) Approximations and inequalities for moving sums. Methodol Comput Appl Probab 14(3):597–616
Guerriero M, Willett P, Glaz J (2009) Distributed target detection in sensor networks using scan statistics. IEEE Trans Signal Process 57(7):2629–2639
Haiman G (1999) First passage time for some stationary processes. Stochastic Process Their Appl 80(2):231–248
Haiman G (2007) Estimating the distribution of one-dimensional discrete scan statistics viewed as extremes of 1-dependent stationary sequences. J Stat Planning Inference 137(3):821–828
Haiman G, Preda C (2006) Estimation for the distribution of two-dimensional discrete scan statistics. Methodol Comput Appl Probab 8(3):373–382
Kulldorff M (1997) A spatial scan statistic. Commun Stat-Theory Methods 26 (6):1481–1496
Malinowski J, Preuss W (1995) Reliability of circular consecutively-connected systems with multistate components. IEEE Trans Reliab 44(3):532–534
Peng CH (2009) Maxima of moving sums in a Poisson random field. Adv Appl Probab 41(3):647–663
Wang X, Glaz J (2014) Variable window scan statistics for normal data. Commun Stat-Theory Methods 43(10-12):2489–2504
Acknowledgments
We would like to thank the Associate Editor and the referees for helpful suggestions that improved the presentation of the results in this article.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Wu, Q., Glaz, J. Robust Scan Statistics for Detecting a Local Change in Population Mean for Normal Data. Methodol Comput Appl Probab 21, 295–314 (2019). https://doi.org/10.1007/s11009-018-9668-6
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11009-018-9668-6
Keywords
- Minimum p-value statistic
- Moving mean
- Moving median
- Multiple window scan statistic
- Robust scan statistic