Comparison of various statistical methods for detecting disease outbreaks

Byeong Yeob Choi, Ho Kim, Un Yeong Go, Jong Hyeon Jeong, Jae Won Lee

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

In this article, we compared seven statistical methods for detecting outbreaks of infectious disease; Historical limits, English model, SPOTv2, CuSums, Bayesian predictive model, RKI method and Serfling model. We used simulated data and real data to compare those seven methods. Simulated data have parameters such as trend, seasonality, mean and standard deviation. Among these methods, SPOTv2 shows the best performance with a balance between sensitivity and positive predictive value and short time lag. But in datasets having strong trends, Bayesian predictive model, English model and Serfling model perform better than SPOTv2. These methods are also compared through real numerical example.

Original languageEnglish (US)
Pages (from-to)603-617
Number of pages15
JournalComputational Statistics
Volume25
Issue number4
DOIs
StatePublished - Dec 2010

Keywords

  • Infectious disease
  • Missing rate
  • Outbreak
  • Positive predictive value
  • Sensitivity
  • Specificity
  • Time lag

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty
  • Computational Mathematics

Fingerprint Dive into the research topics of 'Comparison of various statistical methods for detecting disease outbreaks'. Together they form a unique fingerprint.

Cite this