Multivariate bayesian latent structure modeling of spatio-temporal health data

• Andrew B. Lawson ; Daniel R. Baer ^[1] ; Jane E. Joseph ^[1]
  1. [1] Medical University of South Carolina

     Medical University of South Carolina

     Estados Unidos

     
• Localización: Proceedings of the 35th International Workshop on Statistical Modelling : July 20-24, 2020 Bilbao, Basque Country, Spain / Itziar Irigoien Garbizu (ed. lit.), Dae-Jin Lee (ed. lit.), Joaquín Martínez Minaya (ed. lit.), María Xosé Rodríguez Álvarez (ed. lit.), 2020, ISBN 978-84-1319-267-3, págs. 148-152
• Idioma: inglés
• Enlaces
  • Texto Completo Libro
• Resumen

  • Spatio-temporal health data is now routinely available. Often when time augments space, the focus is on modeling global spatio-temporal e ects.

    However, temporal e ects are often localized spatially and so it could be important to disaggregate these e ects. This leads to spatial clustering of temporal e ects. Often this disaggregation is approached via latent mixture component models. Extending this approach to multiple disease incidence is the focus of this presentation. The speci c example that is explored, and motivates the detailed modeling, is incidence of mild cognitive impairment (MCI) and Alzheimers disease (AD). MCI is considered a pre-cursor of AD and so there is a temporal latent link between these outcomes. Our models address latent component mixtures for each disease but also coupled components shared between diseases. A case study in annual county level incidence in South Carolina is presented.