Detecção de clusters espacias e espaço-temporais em modelos com excesso de zeros e sobredispersão

Abstract

The Spatial Scan Statistic is one of the most important methods for detecting and monitoring spatial disease clusters. Usually it is assumed that disease cases follow a Poisson or Binomial distribution. In practice, however, case count datasets frequently present na excess of zeroes and/or overdispersion, resulting in the violation of those commonly used models, increasing type I error occurrence. This thesis describes a modi_cation of the Spatial Scan Statistic with the Zero Inated Double Poisson (ZIDP) model to reduce type I error, accommodating simultaneously an excess of zeroes and overdispersion. The null and alternative model parameters are estimated by the Expectation-Maximization algorithm and the p-value is obtained through the Fast Double Bootstrap Test. An application is presented for Hanseniasis data in the Brazilian Amazon. An extension of this statistic in prospective space-time surveillance systems has been studied and in assess their performance Monte Carlo simulations were used.