Assessing conditional independence for log-linear poisson models with random effects

Peter X K Song*, Wenxin Jiang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

In the context of regression models with random effects, repeated responses are traditionally assumed to be mutually independent conditional on the random effects. In order to assess the validity of such an assumption and its impact on parameter inference, we propose an estimating equation methodology where both random effects and within-subject correlation are modeled. This allows a subsequent analysis on the statistical significance of the conditional correlation. We illustrate this method with the epilepsy data of Thall and Vail (1990), and find our method useful in achieving a proper representation for the random effect modeling.

Original languageEnglish (US)
Pages (from-to)1233-1245
Number of pages13
JournalCommunications in Statistics - Theory and Methods
Volume29
Issue number5-6
StatePublished - Dec 1 2000

Keywords

  • Conditional correlation
  • Conditional independence
  • Estimating equations
  • Internal correlation
  • Log-linear models
  • Longitudinal data
  • Random effects

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Statistics and Probability

Fingerprint Dive into the research topics of 'Assessing conditional independence for log-linear poisson models with random effects'. Together they form a unique fingerprint.

Cite this