A comparison of exponentially weighted moving average-based methods for monitoring increases in incidence rate with varying population size

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

12 Scopus Citations
View graph of relations

Author(s)

  • Lianjie SHU
  • Yan SU
  • Wei JIANG
  • Kwok-Leung TSUI

Detail(s)

Original languageEnglish
Pages (from-to)798-812
Journal / PublicationIIE Transactions (Institute of Industrial Engineers)
Volume46
Issue number8
Online published1 May 2014
Publication statusPublished - 2014

Abstract

Estimation of incidence rate and quick detection of its increases are important tasks in public health surveillance. In addition to being an efficient tool for online parameter estimation, the Exponentially Weighted Moving Average (EWMA) method has been widely used as an effective monitoring tool in statistical process control. Motivated by its successful applications, several EWMA-type methods are discussed for monitoring and estimating the incidence rate of adverse events in health care applications. The comparison results show that the conventional EWMA chart has a superior performance in detecting small shifts that occur at the start-up but very poor performance when shifts occur at a later time point. Instead, the adaptive EWMA method that is capable of dynamically updating its smoothing parameter can provide an overall good detection performance when shifts occur at both the first time point and a later time point. This result is validated using male thyroid cancer data in New Mexico. Copyright © IIE.

Research Area(s)

  • Inertia, Poisson distribution, quick detection, robust statistics, statistical process control, surveillance

Citation Format(s)

A comparison of exponentially weighted moving average-based methods for monitoring increases in incidence rate with varying population size. / SHU, Lianjie; SU, Yan; JIANG, Wei; TSUI, Kwok-Leung.

In: IIE Transactions (Institute of Industrial Engineers), Vol. 46, No. 8, 2014, p. 798-812.

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review