Bayesian modeling of temperature-related mortality with latent functional relationships

It is common for the mortality rate to increase during periods of extreme temperature and for the minimum mortality rate to depend on factors such as the mean summer temperature. In this paper, local correlation is explicitly described using a generalized additive model with a spatial component which allows information from neighbouring locations to be combined. Random walk and random field models are proposed to describe temporal and spatial correlation structure. Further, joint spatial-temporal modeling is proposed by including a temperature-related mortality term. This will make use of existing data more efficiently and should reduce prediction variability. The methods are illustrated using simulated data based on real mortality and temperature data.