Robustness of the BYM model in absence of spatial variation in the residuals
Open Access
- 1 January 2007
- journal article
- Published by Springer Nature in International Journal of Health Geographics
- Vol. 6 (1) , 39
Abstract
In the context of ecological studies, the Bayesian hierarchical Poisson model is of prime interest when studying the association between environmental exposure and rare diseases. However, adding spatially structured extra-variability in the model fitted to the data when such extra-variability does not exist conditionally on the covariates included in the model (over-fitting) may bias the estimation of the ecological association between covariates and relative risks toward the null. In order to investigate that possibility, a simulation study of the impact of introducing unnecessary residual spatial structure in the estimation model was conducted.Keywords
This publication has 8 references indexed in Scilit:
- Evaluation of Bayesian models for focused clustering in health dataEnvironmetrics, 2007
- Ecological association between indoor radon concentration and childhood leukaemia incidence in France, 1990–1998European Journal Of Cancer Prevention, 2005
- Incidence of childhood leukaemia and non-Hodgkinʼs lymphoma in France: National Registry of Childhood Leukaemia and Lymphoma, 1990–1999European Journal Of Cancer Prevention, 2004
- Sensitivity Analyses for Ecological RegressionBiometrics, 2003
- Disease mapping models: an empirical evaluationStatistics in Medicine, 2000
- Spatial Correlation in Ecological AnalysisInternational Journal of Epidemiology, 1993
- Bayesian image restoration, with two applications in spatial statisticsAnnals of the Institute of Statistical Mathematics, 1991
- Notes on continuous stochastic phenomena.1950