A graphical case statistic for assessing posterior influenceo

Robert E. Weiss, R. Dennis Cook

Research output: Contribution to journalArticlepeer-review

32 Scopus citations

Abstract

SUMMARY: A single number summary of the influence of an observation on the posterior of the parameter of interest is useful, but does not convey all aspects of the effect of deleting the observation. We propose a plot that does convey all the effects of an observation on the posterior.

Original languageEnglish (US)
Pages (from-to)51-55
Number of pages5
JournalBiometrika
Volume79
Issue number1
DOIs
StatePublished - Mar 1 1992

Keywords

  • Bayes's theorem
  • Bayesian data analysis
  • Dynamic graphics
  • Influential observations

Fingerprint

Dive into the research topics of 'A graphical case statistic for assessing posterior influenceo'. Together they form a unique fingerprint.

Cite this