Bayesian spatio-temporal conditional overdispersion models proposals

  1. Mabel Morales-Otero 1
  2. Vicente Núñez-Antón 2
  1. 1 Universidad de Navarra
    info

    Universidad de Navarra

    Pamplona, España

    ROR https://ror.org/02rxc7m23

  2. 2 Universidad del País Vasco/Euskal Herriko Unibertsitatea
    info

    Universidad del País Vasco/Euskal Herriko Unibertsitatea

    Lejona, España

    ROR https://ror.org/000xsnr85

Konferenzberichte:
Proceedings of the 37th International Workshop on Statistical Modelling: July 17-21, 2022 Dortmund, Germany

Verlag: TU Dortmund University

ISBN: 978-3-947323-42-5

Datum der Publikation: 2023

Seiten: 237-242

Art: Konferenz-Beitrag

GOOGLE SCHOLAR

Zusammenfassung

In this work, we proposea direct spatio-temporal extension of the spatial conditional over dispersion models, where we include the spatial lag of the response variable for each time unit in thel inear predictor. The proposed models are able to capture both spatial and temporal correlations that may be present in the data under study. In addition, we also propose temporally varying spatial lag coefficient models, which allow us to study the variation in time of the spatial term. In order to illustrate their performance, we apply our proposals, for Poisson distributed responses, to the Glasgow respiratory hospital admissions dataset, where we compare their performance with the widely used Knorr-Held’ smodels.