Sistemas prospectivos para vigilância espaço-tempo

• Ano de defesa: 2011
• Autor(a) principal: Thais Rotsen Correa
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Estatística; Analise multivariada; Analise por conglomerados; Análise espacial (Estatística)
• Link de acesso: http://hdl.handle.net/1843/BUOS-95ZJYL

The demand for systems capable of detecting changes in spatial and temporal patterns of occurrence of events has grown in several areas of the knowledge. Technological advances in recent years have facilitated the collection and analysis of geographic information, causing an increase in interest in systems for detecting space-time clusters. The study of this type of system is still new and there are few proposals. In this thesis we study systems of prospective space-time surveillance using point data and areal data. We propose two systems, one for point data and one for areal data, both based on the Shiyayaev-Roberts statistic. The efficiency of these systems is evaluated via simulation. Real datasets are used to illustrate the application of these systems. We propose three other systems, with similar properties, for point data. Simulations are also used to check the performance of these systems. Finally, we make a critical analysis about the use of the scan statistic in the prospective context.