Un Modelo jerárquico Bayesiano espacio-temporal con variable de conteos; aplicación de VIH/SIDA en Costa Rica

Abstract

Spatial models that smooth standardized mortality ratios are widely used in di-sease mapping. Usually, estimation is imprecise when events are rare. In situations where each areal count splits into different time periods, this problem is more evi-dent because of the presence of even lower counts for the areal units for each timeperiod. In this work, we analyze models that include geographic and temporal in-formation and some covariates such as percentage of urban household, percentageof people between 24 and 49 years old, and infant mortality ratio of each countyin 2011. As a result, these models produce better estimations, especially for themodel with the simplest space-time interaction. Finally, HIV/AIDS mortality datain Costa Rica (1998-2012) are used as an illustration to compare classic standar-dized mortality ratios and posterior means of relative risk. The proposed methodis more efficient and more precise than the maximum likelihood.