Analysis of double-peak seasonality in the aetiology of perinatal mortality and childhood acute lymphoblastic leukaemia using the Walter-Elwood method

T. A. Nyári, K. Virág, R. J.Q. Mcnally

Research output: Contribution to journalArticle

Abstract

Our study demonstrates the use of the Walter-Elwood method in double-peak seasonal variation. The concept of the geometrical model for analysing cyclic variation is described. Monte Carlo simulation procedures are used to compare the performance of the Walter-Elwood and negative binomial regression methods with double-peak seasonality, in both a comparison between the two methods and a power analysis. The results of 10,000 independent Monte Carlo simulations showed that the Walter-Elwood method and the negative binomial regression analysis identified the same peak in 9,956 samples, indicating that the power of both methods is similar in analysing double-peak cyclic trends. Additionally, two epidemiological applications of double-peak seasonality are presented, which were analysed using the Walter-Elwood method. Further, this is the first study to describe the power of the Walter-Elwood method for double peak seasonality. In conclusion, double-peak seasonality could be investigated with the Walter-Elwood method in ecological studies when only the population at risk is available and there is no other variable.

Original languageEnglish
Pages (from-to)3941-3948
Number of pages8
JournalApplied Ecology and Environmental Research
Volume17
Issue number2
DOIs
Publication statusPublished - 2019

Keywords

  • Childhood leukaemia introduction
  • Cyclic variation
  • Environmental effect
  • Geometrical model
  • Monte-carlo simulations
  • Perinatal mortality
  • Population at risk
  • Power analyses

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Agronomy and Crop Science

Fingerprint Dive into the research topics of 'Analysis of double-peak seasonality in the aetiology of perinatal mortality and childhood acute lymphoblastic leukaemia using the Walter-Elwood method'. Together they form a unique fingerprint.

  • Cite this