Expectation-based scan statistics for monitoring spatial time series data

Research output: Contribution to journalArticle

Abstract

We consider the simultaneous monitoring of a large number of spatially localized time series in order to detect emerging spatial patterns. For example, in disease surveillance, we detect emerging outbreaks by monitoring electronically available public health data, e.g. aggregate daily counts of Emergency Department visits. We propose a two-step approach based on the expectation-based scan statistic: we first compute the expected count for each recent day for each spatial location, then find spatial regions (groups of nearby locations) where the recent counts are significantly higher than expected. By aggregating information across multiple time series rather than monitoring each series separately, we can improve the timeliness, accuracy, and spatial resolution of detection. We evaluate several variants of the expectation-based scan statistic on the disease surveillance task (using synthetic outbreaks injected into real-world hospital Emergency Department data), and draw conclusions about which models and methods are most appropriate for which surveillance tasks.

Original languageEnglish (US)
Pages (from-to)498-517
Number of pages20
JournalInternational Journal of Forecasting
Volume25
Issue number3
DOIs
StatePublished - Jul 1 2009

Fingerprint

Time series data
Statistics
Monitoring
Surveillance
Emergency department
Timeliness
Public health
Aggregate data
Multiple time series

Keywords

  • Biosurveillance
  • Event detection
  • Pattern detection
  • Spatial scan statistics
  • Time series monitoring

ASJC Scopus subject areas

  • Business and International Management

Cite this

Expectation-based scan statistics for monitoring spatial time series data. / Neill, Daniel.

In: International Journal of Forecasting, Vol. 25, No. 3, 01.07.2009, p. 498-517.

Research output: Contribution to journalArticle

@article{08eb4ef452eb4ebeb1e92ece9136d262,
title = "Expectation-based scan statistics for monitoring spatial time series data",
abstract = "We consider the simultaneous monitoring of a large number of spatially localized time series in order to detect emerging spatial patterns. For example, in disease surveillance, we detect emerging outbreaks by monitoring electronically available public health data, e.g. aggregate daily counts of Emergency Department visits. We propose a two-step approach based on the expectation-based scan statistic: we first compute the expected count for each recent day for each spatial location, then find spatial regions (groups of nearby locations) where the recent counts are significantly higher than expected. By aggregating information across multiple time series rather than monitoring each series separately, we can improve the timeliness, accuracy, and spatial resolution of detection. We evaluate several variants of the expectation-based scan statistic on the disease surveillance task (using synthetic outbreaks injected into real-world hospital Emergency Department data), and draw conclusions about which models and methods are most appropriate for which surveillance tasks.",
keywords = "Biosurveillance, Event detection, Pattern detection, Spatial scan statistics, Time series monitoring",
author = "Daniel Neill",
year = "2009",
month = "7",
day = "1",
doi = "10.1016/j.ijforecast.2008.12.002",
language = "English (US)",
volume = "25",
pages = "498--517",
journal = "International Journal of Forecasting",
issn = "0169-2070",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Expectation-based scan statistics for monitoring spatial time series data

AU - Neill, Daniel

PY - 2009/7/1

Y1 - 2009/7/1

N2 - We consider the simultaneous monitoring of a large number of spatially localized time series in order to detect emerging spatial patterns. For example, in disease surveillance, we detect emerging outbreaks by monitoring electronically available public health data, e.g. aggregate daily counts of Emergency Department visits. We propose a two-step approach based on the expectation-based scan statistic: we first compute the expected count for each recent day for each spatial location, then find spatial regions (groups of nearby locations) where the recent counts are significantly higher than expected. By aggregating information across multiple time series rather than monitoring each series separately, we can improve the timeliness, accuracy, and spatial resolution of detection. We evaluate several variants of the expectation-based scan statistic on the disease surveillance task (using synthetic outbreaks injected into real-world hospital Emergency Department data), and draw conclusions about which models and methods are most appropriate for which surveillance tasks.

AB - We consider the simultaneous monitoring of a large number of spatially localized time series in order to detect emerging spatial patterns. For example, in disease surveillance, we detect emerging outbreaks by monitoring electronically available public health data, e.g. aggregate daily counts of Emergency Department visits. We propose a two-step approach based on the expectation-based scan statistic: we first compute the expected count for each recent day for each spatial location, then find spatial regions (groups of nearby locations) where the recent counts are significantly higher than expected. By aggregating information across multiple time series rather than monitoring each series separately, we can improve the timeliness, accuracy, and spatial resolution of detection. We evaluate several variants of the expectation-based scan statistic on the disease surveillance task (using synthetic outbreaks injected into real-world hospital Emergency Department data), and draw conclusions about which models and methods are most appropriate for which surveillance tasks.

KW - Biosurveillance

KW - Event detection

KW - Pattern detection

KW - Spatial scan statistics

KW - Time series monitoring

UR - http://www.scopus.com/inward/record.url?scp=67449160454&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67449160454&partnerID=8YFLogxK

U2 - 10.1016/j.ijforecast.2008.12.002

DO - 10.1016/j.ijforecast.2008.12.002

M3 - Article

AN - SCOPUS:67449160454

VL - 25

SP - 498

EP - 517

JO - International Journal of Forecasting

JF - International Journal of Forecasting

SN - 0169-2070

IS - 3

ER -