A problem in the Bayesian analysis of data without gold standards

Nick Gray, Marco De Angelis, Dominic Calleja, Scott Ferson

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

Abstract

We review methods of calculating the positive predictive value of a test (the probability of having a condition given a positive test for it) in situations where there is no 'gold standard' way to determine the true classification. We show that Bayesian methods lead to illogical results and instead show that a new approach using imprecise probabilities is logically consistent.
Original languageEnglish
Title of host publicationProceedings of the 29th European Safety and Reliability Conference
EditorsMichael Beer, Enrico Zio
Place of PublicationSingapore
Pages2628-2634
Number of pages7
ISBN (Electronic)9789811127243
DOIs
Publication statusPublished - 26 Sept 2019
Event29th European Safety and Reliability Conference, ESREL 2019 - Hannover, Germany
Duration: 22 Sept 201926 Sept 2019

Conference

Conference29th European Safety and Reliability Conference, ESREL 2019
Country/TerritoryGermany
CityHannover
Period22/09/1926/09/19

Keywords

  • diagnostics
  • Bayes’ rule
  • false positives
  • prevalence
  • sensitivity
  • specificity
  • uncertainty
  • gold standard

Fingerprint

Dive into the research topics of 'A problem in the Bayesian analysis of data without gold standards'. Together they form a unique fingerprint.

Cite this