A dependência espacial para análise de dados de demanda por transportes, que está entre as principais questões analítico-espaciais consideradas na análise de transportes, constituiu o foco deste trabalho. Ignorar questões de análise espacial pode invalidar os resultados da análise, levar a previsões inadequadas e, conseqüentemente, a um planejamento ineficiente. Em virtude disso, admitiu-se que a introdução de indicadores de dependência espacial na modelagem de demanda por transportes deveria produzir resultados mais precisos e, desta forma, mais confiáveis dos que os obtidos com modelos tradicionais. Neste sentido, o principal objetivo deste trabalho foi comparar a projeção de demanda por transportes, especificamente na fase de previsão de viagens produzidas de base domiciliar, realizada através de modelos convencionais e de modelos alternativos, que introduzem indicadores para medir a dependência espacial. O trabalho é todo desenvolvido em ambiente SIG (Sistemas de Informações Geográficas), através de ferramentas de análise e estatística espacial, assim como ferramentas de planejamento de transportes de um SIG-T (SIG para Transportes). As ferramentas de análises espaciais serviram tanto para produzir os indicadores de dependência espacial (variáveis espaciais) como para avaliar os resultados dos modelos. Aplica-se o método, que avalia a introdução de indicadores globais e locais de dependência espacial nos modelos alternativos, através de um estudo de caso na cidade de Porto Alegre - RS, que tem por base dados de pesquisa de origem e destino (O-D) obtidos através de entrevista domiciliar (EDOM) em dois períodos distintos (1974 e 1986). Estas informações correspondem aos dados necessários do ano base, que foram utilizados na calibração dos modelos, e do ano meta, que constituíram as informações necessárias para análise dos resultados de estimativas futuras. Conclui-se que a introdução de variáveis espaciais é importante, uma vez que os melhores resultados foram obtidos com modelos alternativos, tanto na etapa de calibração e diagnóstico dos modelos como na etapa de validação (estimativas futuras). No entanto, a dinâmica apresentada pelo desenvolvimento urbano, como é o caso de Porto Alegre, acarreta alterações nas relações entre as diferentes variáveis com o fenômeno estudado, modificando, inclusive, os padrões espaciais. Esta conclusão é dada pelo fato que, o modelo mais ajustado para os dados do ano base não foi o que apresentou os melhores resultados para estimativas futuras. Isto conduz à hipótese, a ser explorada em trabalhos futuros, de que a análise desta dinâmica e o estudo de formas de considerá-la nos modelos de demanda por transportes pode produzir resultados ainda melhores / The degree of spatial data dependence, which is among the issues of spatial analysis that should be considered in transportation planning, is the focus of this study. Ignoring this particularity of data can: produce wrong estimates; jeopardize the results of analyses; and, as a consequence, lead to unsuccessful planning. Therefore, the basic assumption of this work was that the inclusion of spatial dependence indicators can produce more accurate and reliable estimates than those obtained with traditional model structures. In order to test this hypothesis, the main objective of this study was to compare demand predictions produced by traditional models with those produced by alternative models that include indicators of spatial dependence. The study was limited to home-based production trip models, which are part of the trip generation phase of the traditional four-step modeling approach. All work was conducted in a GIS (Geographic Information System) environment, making use of spatial statistics and analysis tools, as well as transportation planning tools available in a GIS-T (i.e., a dedicated GIS for Transportation). Spatial analyses tools were used to generate the spatial dependence indicators and to evaluate the results of the application. A case study was carried out in the city of Porto Alegre, which is the capital of the brazilian state of Rio Grande do Sul, for evaluating the impacts of the addition of global and local indicators of spatial dependence in the models. Two O-D surveys carried out in the years 1974 and 1986 provided the data needed for calibration and validation. The first one was taken as the base year and the second one as the goal year. The results of the application showed that the performance of the models can be improved in both calibration and validation phases with the insertion of spatial variables. However, the urban growth observed in a very dynamic context, such as in the city studied, may dramatically change the relationships between variables, including their spatial patterns. That aspect was responsible for the fact that the model with the best performance in the calibration phase was not the one producing the most accurate forecasts. It raised the hypothesis, to be explored in future research, that the analysis of those dynamic processes and their consideration into transportation demand models are also needed to improve even further the performance of the models

The study employed Geographical Information System (GIS) technology to investigate the vulnerability of groundwater to NO3 content in Buncombe County, North Carolina in two different approaches. In the first study, the spatial distribution of NO3 contamination was analyzed in a GIS environment using Kriging Interpolation. Cokriging interpolation was used to establish how NO3 relates to land cover types and depth to water table of wells in the county. The second study used DRASTIC model to assess the vulnerability of groundwater in Buncombe County to NO3 contamination. To get an accurate vulnerability index, the DRASTIC parameters were modified to fit the hydrogeological settings of the county. A final vulnerability map was created using regression based DRASTIC, a statistic method to measure how NO3 relates to each of the DRASTIC variables. Although the NO3 concentration in the county didn’t exceed the USEPA standard limit (10mg/L), some areas had NO3 as high as 8.5mg/L.

Oral clefts are the most common form of birth defects in the United States (US) and the State of Utah has among the highest prevalence of oral clefts in the nation. The overall objective of this dissertation was to examine the spatial distribution of oral clefts and their linkage with a broad range of demographic, behavioral, social, economic, and environmental risk factors through the application of Geographic Information Systems (GIS) and spatial statistics. Using innovative linked micromaps plots, we investigated the geographic patterns of oral clefts occurrence from 1998 to 2002 and their relationships with maternal smoking rates and proportion of American Indians and Alaskan Natives (AIAN) at large scales across the US. The findings indicated higher oral clefts occurrence in the southwest and the midwest and lower occurrence in the east. Furthermore, these spatial patterns were significantly related to the smoking rates and AIAN. Then at the small area level, hierarchical Bayesian models were built to examine the spatial variation in oral clefts risk in the State of Utah from 1995 to 2004 and to assess association with mothers using tobacco, mothers consuming alcohol during pregnancy, and the proportion of mothers with no high school diploma. Next, multi-scalar spatial clustering and cluster techniques were used to test the hypothesis whether there was spatial clustering of oral clefts anywhere in the State of Utah and whether there were statistically significant local clusters with elevated oral cleft cases. Results generally revealed modest spatial variation in oral clefts risk in the State of Utah, with no pronounced spatial clustering, indicating environmental exposures are unlikely plausible cause of oral clefts. However, a few notable areas within Tri-County Local Health District, Provo/Brigham Young University, and North Orem had a tendency toward elevated oral clefts cases. Investigation of the maternal characteristics of these potential clusters supports the hypotheses that maternal smoking, lower education level, and family history are possible causes of oral clefts. Throughout this dissertation, we demonstrated how birth defects data collected by state and local surveillance systems coupled with GIS and spatial statistics methods can be useful in exploratory etiologic research of birth defects.

Ce manuscrit est dédié à l'étude de l'estimation paramétrique d'une famille de processus ponctuels appelée processus déterminantaux. Ces processus sont utilisés afin de générer et modéliser des configurations de points possédant de la dépendance négative, dans le sens où les points ont tendance à se repousser entre eux. Plus précisément, nous étudions les propriétés asymptotiques de divers estimateurs classiques de processus déterminantaux paramétriques, stationnaires et non-stationnaires, dans les cas où l'on observe une unique réalisation d'un tel processus sur une fenêtre bornée. Ici, l'asymptotique se fait sur la taille de la fenêtre et donc, indirectement, sur le nombre de points observés. Dans une première partie, nous montrons un théorème limite central pour une classe générale de statistiques sur les processus déterminantaux. Dans une seconde partie, nous montrons une inégalité de béta-mélange générale pour les processus ponctuels que nous appliquons ensuite aux processus déterminantaux. Dans une troisième partie, nous appliquons le théorème limite central obtenu à la première partie à une classe générale de fonctions estimantes basées sur des méthodes de moments. Finalement, dans la dernière partie, nous étudions le comportement asymptotique du maximum de vraisemblance des processus déterminantaux. Nous donnons une approximation asymptotique de la log-vraisemblance qui est calculable numériquement et nous étudions la consistance de son maximum. / This manuscript is devoted to the study of parametric estimation of a point process family called determinantal point processes. These point processes are used to generate and model point patterns with negative dependency, meaning that the points tend to repel each other. More precisely, we study the asymptotic properties of various classical parametric estimators of determinantal point processes, stationary and non stationary, when considering that we observe a unique realization of such a point process on a bounded window. In this case, the asymptotic is done on the size of the window and therefore, indirectly, on the number of observed points. In the first chapter, we prove a central limit theorem for a wide class of statistics on determinantal point processes. In the second chapter, we show a general beta-mixing inequality for point processes and apply our result to the determinantal case. In the third chapter, we apply the central limit theorem showed in the first chapter to a wide class of moment-based estimating functions. Finally, in the last chapter, we study the asymptotic behaviour of the maximum likelihood estimator of determinantal point processes. We give an asymptotic approximation of the log-likelihood that is computationally tractable and we study the consistency of its maximum.

Attention deficit/hyperactivity disorder (ADHD) is a common disease with a worldwide prevalence of 5% on preschool children. It has been reported that ADHD patients have volume variant in partial brain regions. Futhermore, functional magnetic resonance imaging have also been used to detect function variant possibility in particular brain regions. In the last decade, some researchers used diffusion MR imaging to investigate the abnormality of neural fibers in disease involved with central nervous system. In general, the diffusion anisotropy of white matter in both ADHD patients and healthy subjects can be estimated seperately to undergo inter-subject comparison. While previous studies often used the popular ICBM brain template (MNI152), this study applied a self-made template of Taiwan adolescents as the common space of image normailization. In this work, group comparison of diffusion fractional anisotropy was performing by using two methods, TBSS and VBM. Both manners found a decreased FA in white matter of ADHD subjects compared with normal control group. However, regions detected by different methods showed low reproducibility. The areas of significant difference include inferior longitudinal fasciculus¡Binternal capsule¡Bexternal capsule¡Bsuperior longitudinal fasciculus¡Boptic radiation¡Bsuperior frontal¡Bsuperior region of corona radiata¡Bcorticospinal tract¡Bposterior region of corona radiata / superior longitudinal fasciculus¡Bsuperior fronto-occipital fasciculus¡Banterior region of corona radiata¡Bgenu of corpus callosum nerve fibers.

Attention deficit hyperactivity disorder (ADHD) is a neurobehavior developmental disorder that affects around 7.5% of Taiwan children. With the use of magnetic resonance imaging , many results have been reported that ADHD patients have volume atrophy in gray matter and dysfunction in couples of cortical regions. In recent years, diffusion MR imaging with diffusion-sensitizing gradients has been used to investigate the abnormality of neural fibers in disease involved with central nervous system. In this study, the anisotropy of white matter in both ADHD patients and age-matched healthy subjects was estimated using diffusion tensor imaging to undergo inter-subject comparison. In this work, a significant decrease (FWE-corrected p-value <0.05) of FA values has been found in white matter of adolescents diagnosed as ADHD patients, compared with normal controls group. The areas that confirmed by two different algorithms of inter-subject comparison are mainly diffused on white matter region, including middle cerebellar peduncle, left inferior longitudinal fasciculus, internal capsule, left optic radiation, external capsule, splenium of the corpus callosum, superior longitudinal fasciculus, superior frontal and parietal-occipital nerve fibers.

No description available.

Parmi les modèles permettant d'introduire de l'interaction entre les points, nous trouvons très large famille des modèles gibbsiens de processus ponctuels spatiaux issus de la physique statistique, permettant de modéliser à la fois des motifs répulsifs ou attractifs. Dans cette thèse, nous nous intéressons à l'inférence semi-paramétrique de ces modèles caractérisés par l'intensité conditionnelle de Papangelou. Deux contextes sont étudiés. Dans le premier thème, nous décrivons une procédure d'estimation du terme d'interaction du premier ordre (qui peut être aussi appelé l'intensité de Poisson) de l'intensité conditionnelle de Papangelou. L'idée sur laquelle l'estimation est basée permet, sous l'hypothèse d'une portée finie, de négliger les termes d'interaction d'ordre supérieur quelle que soit leur nature. La consistance forte et la normalité asymptotique de l'estimateur sont prouvées. Une étude par simulations illustre la performance de l'estimateur sur une fenêtre d'observation finie. Dans le second thème, nous nous focalisons sur la classe la plus connue et utilisée; le processus ponctuel à interaction par paires. Nous construisons une nouvelle méthode d'estimation de la fonction d'interaction de paires dans l'esprit des estimations non paramétriques par lissage à partir d'une réalisation du processus ponctuel spatial à interaction par paires. Deux cas sont étudiées: le cas stationnaire et le cas isotrope. Ces estimateurs exploitent à nouveau la propriété de portée finie des processus ponctuels et intégrent l'estimation du paramètre de l'intensité de Poisson vue dans le premier thème. Nous présentons les propriétés asymptotiques telles que la consistance forte ponctuelle, la consistance forte globale avec différentes vitesses de consistance, le comportement de l'erreur quadratique moyenne et la normalité asymptotique de ces estimateurs. / Among models allowing to introduce interaction between points, we find the large class of Gibbs models coming from statistical physics. Such models can produce repulsive as well as attractive point pattern. In this thesis, we are interested in the semi-parametric inference of such models characterized by the Papangelou conditional intensity. Two frameworks are considered. First, we describe a procédure which intends to estimate the first-order interaction term (also called Poisson intensity) of the Papangelou conditional intensity. Under the assumption of finite range of the process, the idea upon which the procedure is based allows us to neglect higher-order interaction terms. We study the stong consistency and the asymptotic normality and conduct a simulation study which highlights the efficiency of the method for finite observation window. Second, we focus on the main class of Gibbs models which is the class of pairwise interaction point processes. We construct a kernel-based estimator of the pairwise interaction function. Two cases are studied: the stationary case and the isotropic case.The estimators, we propose, exploit the finite range property and the estimator of the Poisson intensity defined in the first part. We present asymptotic properties, namely the strong consistency, the behavior of the mean squared error and the asymptotic normality.

Com o recente avanço do poder computacional, a amostragem de curvas indexadas espacialmente tem crescido principalmente em dados ecológicos, atmosféricos e ambientais, o que conduziu a adaptação de métodos geoestatísticos para o contexto de Análise de Dados Funcionais. O objetivo deste trabalho é estudar métodos de krigagem para Dados Funcionais, adaptando os métodos de interpolação espacial em Geoestatística. Mais precisamente, em um conjunto de dados funcionais pontualmente fracamente estacionário e isotrópico, desejamos estimar uma curva em um ponto não monitorado no espaço buscando estimadores não viciados com erro quadrático médio mínimo. Apresentamos três abordagens para aproximar uma curva em sítio não monitorado, demonstramos resultados que simplificam o problema de otimização postulado pela busca de estimadores ótimos não viciados, implementamos os modelos em MATLAB usando ondaletas, que é mais adequada para captar comportamentos localizados, e comparamos os três modelos através de estudos de simulação. Ilustramos os métodos através de dois conjuntos de dados reais: um conjunto de dados de temperatura média diária das províncias marítimas do Canadá (New Brunswick, Nova Scotia e Prince Edward Island) coletados em 82 estações no ano 2000 e um conjunto de dados da CETESB (Companhia Ambiental do Estado de São Paulo) referentes ao índice de qualidade de ar MP10 em 22 estações meteorológicas na região metropolitana da cidade de São Paulo coletados no ano de 2014. / The advance of the computational power in last decades has been generating a considerable increase in datasets of spatially indexed curves, mainly in ecological, atmospheric and environmental data, what have leaded to adjustments of geostatistcs for the context of Functional Data Analysis. The goal of this work is to adapt the kriging methods from geostatistcs analysis to the framework of Functional Data Analysis. More precisely, we shall interpolate a curve in an unvisited spot searching for an unbiased estimator with minimum mean square error for a pointwise weakly stationary and isotropic functional dataset. We introduce three different approaches to estimate a curve in an unvisited spot, we demonstrate some results simplifying the optimization problem postulated by the optimality from these estimators, we implement the three models in MATLAB using wavelets and we compare them by simulation. We illustrate the ideas using two dataset: a real climatic dataset from Canadian maritime provinces (New Brunswick, Nova Scotia and Prince Edward Island) sampled at year 2000 in 82 weather station consisting of daily mean temperature and data from CETESB (environmental agency from the state of São Paulo, Brazil) sampled at 22 weather station in the metropolitan region of São Paulo city at year 2014 consisting of the air quality index PM10.

Em um experimento, é comum ocorrerem fatores não controláveis, responsáveis pela heterogeneidade entre as parcelas. Mesmo executando os três princípios básicos da experimentação no planejamento (repetição, casualização e controle local), ainda assim, pode haver correlação nos erros e, portanto, dependência espacial na área estudada. Se for detectada essa estrutura de auto-correlação e se essa informação for utilizada na análise estatística, estimativas mais eficientes dos contrastes entre as médias dos tratamentos são garantidas, mas se tal estrutura for desconsiderada pode impedir que diferenças reais sejam detectadas. Neste trabalho, foram observadas as coordenadas dos centros das parcelas de um delineamento em blocos ao acaso. A variável resposta, deste experimento, é a concentração de carbono orgânico no solo, sendo as avaliações feitas no início do experimento, ou seja, antes da aplicaçao dos tratamentos, portanto, um ensaio em branco, um ano após a aplicação dos tratamentos e, novamente, depois de mais um ano. Para tanto, foram utilizadas as análises clássica e espacial na comparação dos métodos de estimação de contrastes de médias de tratamentos. O método estudado para a análise clássica, em que considera que os erros são não correlacionados, foi o dos mínimos quadrados ordinários. Já para a análise, levando em consideração a dependência espacial, foram utilizados o modelo geoestatístico, em que consiste na adição de um efeito aleatório com correlação, e o modelo de Papadakis, que consiste na adição de uma covariável construída a partir de observações em parcelas vizinhas. No modelo geoestatístico foi verificada a presença da dependência espacial através dos critérios de informação de Akaike e de informação Bayesiano ou de Schwarz e os métodos testados foram o do variograma seguido de mínimos quadrados generalizados e o da máxima verossimilhança. Para o modelo de Papadakis, foi testada a significância da covariável referente duas médias dos resíduos entre as parcelas vizinhas e a própria parcela tanto no modelo em blocos ao acaso quanto no modelo inteiramente casualizado, e o teste não foi significativo em nenhum dos dois casos. Mesmo assim, os cálculos foram realizados para esse método, mostrando que para esse conjunto de dados, este método não é indicado. Fazendo uso de algumas medidas de comparação desses métodos, para os dados em questão, o método de estimação dos contrastes de médias de tratamentos que apresentou as medidas de comparação mais dispersas foi o do modelo de Papadakis e o menos disperso foi o da máxima verossimilhança. Ainda, pelos intervalos de confiança, observou-se que na análise espacial, outros contrastes diferiram de zero significativamente, além daqueles que foram observados na análise clássica, o que se conclui que quando é levada em consideração a autocorrelação dos erros, os contrastes são estimados com maior eficiência / Not controllable factors is common occur in experiments, they are responsible for the heterogeneity among parcels. Even executing the three experimentation basic principles in the design (repetition, randomization and local control), even so, may have correlation in errors and, therefore, spatial dependence in the area of study. If that autocorrelation structure is detected and if this information is used in statistical analysis, estimates more efficient of contrasts among treatments means are guaranteed, but if this structure is disregarded can prevent that real diferences are detected. In this work, the coordinates of parcels centers in a design of random blocks were observed. The concentration of soil organic carbon is the response variable of this experiment, with the available made at the beginning of the experiment, ie, before the treatments application, therefore, a blank, a year after the treatments application and, again, after a year. Then, the classical and spatial analysis were used to compare the methods of estimation of treatments means contrasts. The method studied for the classical analysis, which considers that the errors are not correlated, was the ordinary least squares. For the analysis, considering the spatial dependence, were used the geostatistical model, where consists in the addition of a random effect with correlation, and the Papadakis model, which consists in the addition of a covariate built from observations in neighbouring. In geostatistical model was verified the spatial dependence through the Akaike and Bayesian or Schwarz criteria of information and the methods tested were the variogram followed by generalized least squares and the maximum likelihood. For the Papadakis model, was tested the significance of covariate referring to the average of residuals among neighbouring parcels and own parcel in the random blocks model and in the completely randomized model, and the test was not significant in any of both cases. Still, the calculus were made for this method, showing that for this data set, this method is not indicated. Using some measures to compare these methods, for these data, the method of estimation of treatments means contrasts which presented the measures of comparison more dispersed was the Papadakis model and the less dispersed was the maximum likelihood. Still, in the confidence intervals, it was observed that in spatial analysis other contrasts di®ered from zero significantly, besides of those which were observed in classical analysis, which concludes that when the autocorrelation of errors is considering, the contrasts are estimated with greater e±ciency.