Modelagem espaco-temporal de processos pontuais: aplicação em circuitos de televisão para controle de crimes

• Ano de defesa: 2016
• Autor(a) principal: Ana Gabriela Braga
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: MCMC; Inferência bayesiana; Criminalidade; Processos pontuais espaço-temporais; Estatistica matematica; Criminalidade urbana; Estatistica; Análise de séries temporais
• Link de acesso: http://hdl.handle.net/1843/BUBD-A9ZG44

The increase in population over the centuries has led to signicant impacts on various factors belonging to the everyday life of society, criminality is the principal among them. Therefore, for the authorities, to monitor crimes is indispensable for the creation of public politics to control these infractions. For this reason, the mapping of crimes isnecessary to detect changes on the spatio temporal pattern of the occurrence of these events. The need to introduce strategic solutions that interrupt and revert such factors in a signicant way caused the development of statistical methods to the analysis of processof this nature, mainly due to the vast applicability of the spatio temporal models. In this work, we propose a spatio temporal methodology to point process, which purpose is to map the temporal degradation of the eect of the installation of 60 cameras of Closed Circuit Television (CCTV) on the distribution of crimes against a person, in a region with high criminality of the city Belo Horizonte. This model's inference is done under the Bayesian approach utilizing MCMC methods, more precisely, the Gibbs sampler with Metropolis-Hastings steps.