Orientador: Luiz Koodi Hotta / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática, Estatística e Computação Científica / Made available in DSpace on 2018-08-20T22:42:22Z (GMT). No. of bitstreams: 1 TruciosMaza_CarlosCesar_M.pdf: 13820849 bytes, checksum: 0cc000af0d7cb7cb6ee6c05ef9f3afbd (MD5) Previous issue date: 2012 / Resumo: Mercados financeiros têm mostrado um grande interesse em intervalos de previsão como uma medida de incerteza. Além das previsões do nível, a previsão da volatilidade é importante em várias aplicações em finanças. O modelo GARCH tem sido bastante utilizado na modelagem da volatilidade. A partir deste modelo, outros modelos foram propostos para incorporar outros fatos estilizados, como o efeito de alavancagem. Neste sentido, temos os modelos EGARCH e GJR-GARCH. Os métodos tradicionais de construção de intervalos de previsão para séries temporais geralmente assumem que os parâmetros do modelo são conhecidos e os erros normais. Quando estas suposições não são verdadeiras, o que costuma acontecer na prática, o intervalo de previsão obtido tenderá a ter uma cobertura abaixo da nominal. Nesta dissertação propomos uma adaptação do algoritmo PRR (Pascual, Romo e Ruiz) desenvolvido para obter intervalos de previsão em modelos GARCH, para obter intervalos de previsão em modelos EGARCH e GJR-GARCH. As adaptações feitas são analisadas através de experimentos Monte Carlo e verifica-se que tiveram bom desempenho apresentando valores da cobertura estimada próximos da cobertura nominal. As adaptações propostas assim como o algoritmo PRR são aplicadas para obter intervalos de previsão dos retornos e das volatilidades para a série de retornos da Ibovespa e para a série NYSE COMPOSITE(DJ) da bolsa de valores de Nova Iorque, obtendo em ambos os casos resultados satisfatórios / Abstract: Financial Markets have shown a big interest in forecast intervals (prediction intervals) as a uncertain measure. Besides the level prediction, the prediction of the volatility is very important in many financial applications. The GARCH model, has been very used in volatility modeling. From this model, other have been proposed to incorporate other stylized facts, such as the leverage effect. In this sense, we have the EGARCH and GJR-GARCH models. Traditional methods for constructing predictions intervals for time series generally assume that the model parameters are known and the erros are normal. When these assumptions are not true, that it is very often in practice, the obtained prediction interval, will tend to have a cover under the nominal. In this theses we propose an adaptation of the PRR (Pascual, Romo and Ruiz) algorithm developed to obtain prediction intervals in GARCH models, to obtain prediction intervals in EGARCH and GJR-GARCH models. These adaptations are analized through Monte-Carlo experiments and It was verified that they have a good performance showing estimated cover values close to the nominal cover. The proposed adaptations, such as the PRR algorithm are applied to obtain prediction intervals from the returns and volatilities for the Ibovespa return series and for the New York Stock Exchange NYSE COMPOSITE(DJ) series, obtaining satisfactory results in both cases / Mestrado / Estatistica / Mestre em Estatística

My thesis deals with the very complicated topic of juvenile crime. The circumstances that result in socially pathological phenomenons and crime are now-days relatively easy to happen. The questions of juvenile crime are more frequently discussed not only by the scientists but also by the general public. The goal of my thesis is to outline the problems of the juvenile crime development in the district Klatovy and to use the acquired information in crime prevention and to focus the care on difficult individuals. In the theoretical part, the thesis presents the development of juvenile crime in the Czech Republic and factors that shape and influence it. The practical part of my thesis describes the performed research that is a type of quantitative research from the sociological point of view. The method of questioning, the method of data secondary analysis and the method of model casuistics were used. The targeted group is represented by children and adolescents living in the district Klatovy. The respondents´ principal residence in Klatovy district and the age under 18 are the main criteria of categorization. All respondents were chosen with the method of random stratification. The research was held from January to March of the year 2009. Following hypothesis were determined: Hypothesis 1: The rate of juvenile crime is higher in towns than in villages. Hypothesis 2: Socially pathological features are side-effects of the juvenile crime. Hypothesis 3: Child crime in the district Klatovy is ascending in comparison to adolescent crime in the district Klatovy. I believe that all three hypotheses were confirmed. Based on the OSPOD (Department for juvenile care) reports we can assume that there is na increase in educational problems and subsequently the crime is gradually increasing. If we consider the crime committed by children in the district Klatovy from 2004 to 2008, we find that the year 2007 is a year with the lowest crime rate monitored. We are witnessing the gradually decreasing development trend. Lately, the total development of child crime has been slightly increasing and adolescent crime slowly decreasing.

Orientador: Nancy Lopes Garcia / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática, Estatística e Computação Científica / Made available in DSpace on 2018-08-20T13:18:28Z (GMT). No. of bitstreams: 1 Fadel_DesireeFaria_M.pdf: 12260685 bytes, checksum: 0187ee6ba6eb07c46ce97bd741e51a6c (MD5) Previous issue date: 2012 / Resumo: Neste trabalho, iremos considerar modelos autorregressivos com memória variável estacionários (AR-MV). Em particular, consideraremos modelos cuja memória depende do valor do primeiro antecessor, Yt-1, pertencer a uma partição da reta determinada por um parâmetro 'ALFA' (escalar ou vetorial), chamado de parâmetro limiar. O objetivo deste trabalho é estimar o parâmetro limiar 'ALFA' através de uma adaptação do método proposto por Hansen (2000). A ideia do método é minimizar a soma dos quadrados dos erros estimando, sequencialmente, os coeficientes 'BETA' do modelo autorregressivo (AR) supondo primeiramente que 'ALFA' é conhecido, e depois estimar o parâmetro 'ALFA' utilizando o valor estimado '^BETA' até atingir a convergência. A comparação dos modelos AR-MV com os respectivos AR foi feita através da capacidade de previsão de cada um deles / Abstract: In this paper, we consider stationary autoregressive models with variable memory (AR-MV). In particular, we consider models whose memory depends on the value of the first ancestor, Yt-1, to belong to a partition of the line determined by a parameter 'ALPHA' (scalar or vector), called the threshold parameter. The objective of this study is to estimate the threshold parameter 'ALPHA' by adapting the method proposed by Hansen (2000). The idea of the method is to minimize the sum of squared errors by estimating sequentially the coefficients 'BETA' of the autoregressive model (AR) assuming first that 'ALPHA' is known, and then estimate the parameter 'ALPHA' using the estimated value of '^BETA' until convergence is achieved. The comparison of the AR-MV with the respective AR was performed by the ability to predict each / Mestrado / Estatistica / Mestre em Estatística