Edinburgh Research Explorer

A Comparison of Frequentist and Bayesian Approaches to Latent Class Modelling of Susceptibility to Asthma and Patterns of Antibiotic Prescriptions in Early Life

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

  • Danielle Belgrave
  • Christopher Bishop
  • Adnan Custovic
  • Angela Simpson
  • Aida Semic-Jusufagic
  • Andrew Pickles
  • Iain Buchan

Related Edinburgh Organisations

Open Access permissions


Original languageEnglish
Title of host publicationProceedings of the 26th International Workshop on Statistical Modelling. Valencia (Spain)
Number of pages4
Publication statusPublished - 2011


The assessment of patterns of antibiotic use in early life may have major implications for understanding the development of asthma. This paper compares a classical generalized latent variable modelling framework and a Bayesian machine learning approach to define latent classes of susceptibility to asthma based on patterns of antibiotic use in early life. We compare the potential advantages of each method for elucidating clinically meaningful phenotypes or classes.

ID: 24600709