Abstract
The majority of control charts based on scan statistics for spatio-temporal surveillance use full observation vectors. In high-dimensional applications, dimension-reduction techniques are usually applied. Typically, the dimension reduction is conducted as a post-processing step rather than in the data acquisition stage and thus, a full sample covariance matrix is required. When the dimensionality of data is high, (i) the sample covariance matrix tends to be ill-conditioned due to a limited number of samples; (ii) the inversion of such a sample covariance matrix causes numerical issues; and (iii) aggregating information from all variables may lead to high communication costs in sensor networks. In this paper, we propose a set of reduced-dimension (RD) control charts that perform dimension reduction during the data acquisition process by spatial scanning. The proposed methods avoid computational difficulties and possibly high communication costs. We derive a theoretical measure that characterizes the performance difference between the RD approach and the full observation approach. The numerical results show that the RD approach has little performance loss under several commonly used spatial models while enjoying all the benefits of implementation. A case study on water quality monitoring demonstrates the effectiveness of the proposed methods in real applications.
Similar content being viewed by others
References
Bartram J, Ballance R (1996) Water quality monitoring: a practical guide to the design and implementation of freshwater quality studies and monitoring programmes. CRC Press, New York
Bodnar O, Schmid W (2005) Multivariate control charts based on a projection approach. Allgemeines Statistisches Archiv 89:75–93
Brynjarsdóttir J, Berliner LM (2014) Dimension-reduced modeling of spatio-temporal processes. J Am Stat Assoc 109:1647–1659
Corless RM, Gonnet GH, Hare DE, Jeffrey DJ, Knuth DE (1996) On the LambertW function. Adv Comput Math 5:329–359
Crosier RB (1988) Multivariate generalizations of cumulative sum quality-control schemes. Technometrics 30:291–303
Gaetan C, Guyon X (2010) Spatial statistics and modeling. Springer, New York
Guerriero M, Willett P, Glaz J (2009) Distributed target detection in sensor networks using scan statistics. In: IEEE transactions on signal processing, vol. 57, pp 2629–2639
He Q, Zhou S (2014) Discriminant locality preserving projection chart for statistical monitoring of manufacturing processes. Int J Prod Res 52:5286–5300
Healy JD (1987) A note on multivariate CUSUM procedures. Technometrics 29:409–412
Hotelling, H (1947) “Multivariate quality control,” Techniques of statistical analysis
Jiang W, Han SW, Tsui K-L, Woodall WH (2011) Spatiotemporal surveillance methods in the presence of spatial correlation. Stat Med 30:569–583
Kim S-H, Alexopoulos C, Tsui K-L, Wilson JR (2007) A distribution-free tabular CUSUM chart for autocorrelated data. IIE Trans 39:317–330
Lee J, Hur Y, Kim S-H, Wilson JR (2012) Monitoring nonlinear profiles using a wavelet-based distribution-free CUSUM chart. Int J Prod Res 50:6574–6594
Lee ML, Goldsman D, Kim S-H (2015) Robust distribution-free multivariate CUSUM charts for spatiotemporal biosurveillance in the presence of spatial correlation. IIE Trans Healthc Syst Eng 5:74–88
Lee ML, Goldsman D, Kim S-H, Tsui K-L (2014) Spatiotemporal biosurveillance with spatial clusters: control limit approximation and impact of spatial correlation. IIE Trans 46:813–827
Liu K, Mei Y, Shi J (2015) An adaptive sampling strategy for online high-dimensional process monitoring. Technometrics 57:305–319
Liu K, Zhang R, Mei Y (2016) Scalable SUM-Shrinkage Schemes for Distributed Monitoring Large-Scale Data Streams. arXiv:1603.08652
Lowry CA, Woodall WH, Champ CW, Rigdon SE (1992) A multivariate exponentially weighted moving average control chart. Technometrics 34:46–53
Mishin D, Brantner-Magee K, Czako F, Szalay AS (2014) “Real time change point detection by incremental PCA in large scale sensor data,” In: High performance extreme computing conference (HPEC), 2014 IEEE. IEEE pp 1–6
Page ES (1954) Continuous inspection schemes. Biometrika 41:100–115
Park Y, Baek SH, Kim S-H, Tsui K-L (2014) Statistical process control-based intrusion detection and monitoring. Qual Reliab Eng Int 30:257–273
Ripley BD (2005) Spatial statistics, vol 575. Wiley, New York
Rogerson PA (1997) Surveillance systems for monitoring the development of spatial patterns. Stat Med 16:2081–2093
Rogerson PA, Yamada I (2004) Approaches to syndromic surveillance when data consist of small regional counts. Morbidity and Mortality Weekly Report pp 79–85
Runger GC (1996) Projections and the U-squared multivariate control chart. J Qual Technol 28:313–319
Skubałlska-Rafajlowicz E (2013) Random projections and Hotelling’s T2 statistics for change detection in high-dimensional data analysis. Int J Appl Math Comput Sci 23:447–461
Sonesson C (2007) A CUSUM framework for detection of space–time disease clusters using scan statistics. Stat Med 26:4770–4789
Spiegelhalter D, Sherlaw-Johnson C, Bardsley M, Blunt I, Wood C, Grigg O (2012) Statistical methods for healthcare regulation: rating, screening and surveillance. J R Stat Soc 175:1–47
Tango T (1995) A class of tests for detecting ‘general’and ‘focused’clustering of rare diseases. Stat Med 14:2323–2334
Telci IT, Aral MM (2011) Contaminant source location identification in river networks using water quality monitoring systems for exposure analysis. Water Qual Expos Health 2:205–218
Tsui K-L, Chiu W, Gierlich P, Goldsman D, Liu X, Maschek T (2008) A review of healthcare, public health, and syndromic surveillance. Qual Eng 20:435–450
Ver Hoef JM, Peterson EE (2010) A moving average approach for spatial statistical models of stream networks. J Am Stat Assoc 105:6–18
Wang H, Kim S-H, Huo X, Hur Y, Wilson JR (2015a) Monitoring nonlinear profiles adaptively with a wavelet-based distribution-free CUSUM chart. Int J Prod Res 53:4648–4667
Wang K, Jiang W, Li B (2015b) A spatial variable selection method for monitoring product surface. Int J Prod Res 54:4161–4181
Xie Y, Huang J, Willett R (2013) Change-point detection for high-dimensional time series with missing data. IEEE J Select Top Sign Process. 7:12–27
Xie Y, Siegmund D (2013) Sequential multi-sensor change-point detection. Annal Stat 41:670–692
Xie, Y, Wang M, Thompson A (2015) Sketching for sequential change-point detection. In: 2015 IEEE global conference on signal and information processing (GlobalSIP), IEEE, pp 78–82
Yamamoto M, Hwang H (2017) Dimension-reduced clustering of functional data via subspace separation. J Classif 34:294–326
Zhang F (2006) The Schur complement and its applications, vol 4. Springer, New York
Acknowledgements
This material is based upon work supported by NSF under Grants CMMI-1538746.
Author information
Authors and Affiliations
Corresponding author
Additional information
Handling Editor: Pierre Dutilleul.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Chen, J., Park, C., Kim, SH. et al. To reduce or not to reduce: a study on spatio-temporal surveillance. Environ Ecol Stat 26, 217–238 (2019). https://doi.org/10.1007/s10651-019-00425-4
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10651-019-00425-4