No description available.

This thesis contains theory of probability and statistical sets. Solved and unsolved problems of probability, random variable and distributions random variable, random vector, statistical sets, regression and correlation analysis. Unsolved problems contains solutions.

Submitted by (ana.araujo@ufrpe.br) on 2016-05-25T17:33:53Z No. of bitstreams: 1 Cicero Carlos Ramos de Brito.pdf: 2928725 bytes, checksum: 35b3038da8b53c426368cd7ed2c96e03 (MD5) / Made available in DSpace on 2016-05-25T17:33:53Z (GMT). No. of bitstreams: 1 Cicero Carlos Ramos de Brito.pdf: 2928725 bytes, checksum: 35b3038da8b53c426368cd7ed2c96e03 (MD5) Previous issue date: 2014-08-07 / This work is divided into five chapters. The first one contains the objectives and the relevance of this study. In the second one, we review the literature presenting the state of the art in the field and we give an overview of the most used distributions which are the basis for the ones we generalize in later chapters. In the third chapter, the method for generating distributions is presented by means of a theorem and 7 corollaries. This method extends the probability distribution building process, so that the classes of distributions are constructed from pre-defined univariate monotonic functions and known distributions. In the fourth chapter, similarly to the third one, however, the construction was made from pre-defined multivariate monotonic functions and known multivariate distributions. We also conducted the development of new probability distributions and new generating functions of probability distribution classes. We illustrate the potentiality of this new univariate probability distribution we propose here by means of an application to an actual data set of excesses of flood peaks presented by Choulakian e Stephens (2001). For the application of the new multivariate distribution proposed, we used the database of measurements of Iris Flower exposed in Fisher’s work (1936). Six models were compared and, for their choice, we based on the Akaike Information Criterion (AIC), the Akaike Information Criterion corrected (AICc), Bayesian Information Criterion (BIC), the Hannan Quinn Information Criterion (HQIC) and the statistics of Cramér-von Mises and Anderson-Darling to assess the model fitting. Finally, we present the conclusions from de analyses, the comparisons from the results found in this thesis, the possibilities for research and ways to future works. / Este trabalho divide-se em cinco capítulos. No primeiro trazemos a introdução que contém os objetivos e a relevância deste estudo. No segundo, temos a revisão da literatura em que apresentamos o estado da arte deste campo do conhecimento e fazemos um apanhado das distribuições mais utilizadas que são base para as que generalizamos em capítulos posteriores. No terceiro capítulo, apresenta-se o método gerador, que é um teorema proposto com 7 corolários, que estende o processo de construções de distribuições de probabilidades, a fim de que as classes de distribuições sejam construídas a partir de funções monotônicas univariadas pré-definidas e distribuições conhecidas. No quarto capítulo foi trabalhado semelhantemente ao terceiro, entretanto, a construção se deu a partir das funções monotônicas multivariadas pré-definidas e distribuições multivariadas conhecidas. Também foi realizado o desenvolvimento das novas distribuições probabilísticas e novas funções geradoras de classes de distribuições probabilísticas. Ilustramos a potencialidade da nova distribuição de probabilidade univariada aqui proposta através de uma aplicação ao conjunto de dados reais de excessos de picos de enchentes apresentado em Choulakian e Stephens (2001). Para uma aplicação da nova distribuição multivariada proposta, utilizou-se a base de dados de medidas da Flor de Iris apresentada no trabalho de Fisher (1936). São comparados seis modelos e para a seleção desses modelos, foram utilizados o Critério de Informação de Akaike (AIC), o Critério de Informação de Akaike corrigido (AICc), o Critério de Informação Bayesiano (BIC), o Critério de Informação Hannan Quinn (HQIC) e as estatísticas de Cramer Von-Mises e de Anderson-Darling para avaliar o ajuste dos modelos. Por fim, apresentamos as conclusões a partir das análises e comparações dos resultados obtidos e direções a trabalhos futuros.

Submitted by (ana.araujo@ufrpe.br) on 2016-08-03T15:37:53Z No. of bitstreams: 1 Luiz Medeiros de Araujo Lima Filho.pdf: 6714156 bytes, checksum: 8537b197237b8261b5b264ad769ab466 (MD5) / Made available in DSpace on 2016-08-03T15:37:53Z (GMT). No. of bitstreams: 1 Luiz Medeiros de Araujo Lima Filho.pdf: 6714156 bytes, checksum: 8537b197237b8261b5b264ad769ab466 (MD5) Previous issue date: 2012-10-01 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Knowledge of growth and production present and future of trees in forest stands is a key element in sustainable forest management. It is intended in this thesis to estimate the frequency distribution by diameter class, estimate and compare volumetric equations via symmetric nonlinear models as well, propose to compare models and adjust volume growth via symmetrical nonlinear models in Eucalyptus spp. clones in the region of the Araripe - PE. The experiment was conducted at the Experimental Station of the Agricultural Research Company of Pernambuco (IPA), located in the municipality of Araripina - PE. This experiment consists of 15 clones of Eucalyptus spp., species and hybrids resulting from natural crossing and controlled pollination. The database is composed of all the survival trees of the experiment, corresponding to 1189 trees, divided into three groups. To estimate the diameter distributions were adopted probability distributions Normal, with three parameters Weibull, Log-normal and Gamma. The next step was to fit of the model of Schumacher and Hall and Spurr model associated the following symmetric distributions: Normal, Student t, Exponential Power and Logistics II. The volume growth models in this thesis were adjusted with symmetric distributions associated with them. In the estimation of the diameter distributions, the results were similar in the three groups, with emphasis on the Log-normal distribution. This distribution was the most appropriate for most of the period. Analyzing the volumetric equations, it was found that the model was Schumacher and Hall the most suitable, when associated with the Student t distribution with three degrees .of freedom and Exponential Power (t = 0,8) respectively to groups I and II. In group III, the model Spurr associated with Exponential Power (t = 0,8) distribution had the best performance. Comparing the volume growth models, it was observed that the proposed models best fits obtained in this thesis, when the distributions associated with the Student t with three degrees of freedom and Exponential Power (t = 0,8), respectively, in groups I and III. In group II, the Chapman-Richards model associated with Student t distribution with three degrees of freedom had the best performance. / O conhecimento do crescimento e da produção presente e futura de árvores em povoamentos florestais é elemento fundamental no manejo florestal sustentável. Desta forma, pretende-se nesta tese estimar a distribuição de frequência por classe diamétrica, estimar e comparar equações volumétricas via modelos simétricos não lineares, bem como, propor, ajustar e comparar modelos de crescimento em volume via modelos simétricos não lineares em clones de Eucalyptus spp. na região da Chapada do Araripe - PE. O experimento foi desenvolvido na Estação Experimental da Empresa Pernambucana de Pesquisa Agropecuária (IPA), localizada no município de Araripina - PE. Esse experimento é composto por 15 clones de Eucalyptus spp., resultantes de espécies e híbridos de cruzamento natural e polinização controlada. A base de dados foi composta por todas as árvores sobreviventes do experimento, que corresponde a 1189 árvores, dividida em três grupos. Para estimar as distribuições diamétricas foram adotadas as distribuições de probabilidade Normal, Weibull com três parâmetros, Log-normal e Gama. Em seguida, procedeu-se com os ajustes dos modelos volumétricos de Schumacher e Hall e de Spurr associados as seguintes distribuições simétricos: Normal, t de Student, Exponencial Potência e Logística II. Posteriormente, os modelos de crescimento em volume propostos nesta tese foram ajustados associados às mesmas distribuições simétricas. Na estimativa das distribuições diamétricas, os resultados foram similares nos três grupos, com destaque para a distribuição Log-normal. Essa distribuição foi a mais significativa na maior parte do período avaliado. Analisando as equações volumétricas, observou-se que o modelo de Schumacher e Hall obteve o melhor desempenho, quando associado as distribuições t de Student com três graus de liberdade e Exponencial Potência (t = 0,8), respectivamente, para os grupos I e II. No grupo III, o modelo de Spurr associado a distribuição Exponencial Potência (t = 0,8) obteve o melhor desempenho. Comparando os modelos de crescimento em volume, observou-se que os modelos propostos nesta tese obtiveram melhores ajustes, quando associados as distribuições t de Student com três graus de liberdade e Exponencial Potência (t = 0,8), respectivamente, nos grupos I e III. No grupo II, o modelo de Chapman-Richards associado a distribuição t de Student com três graus de liberdade obteve o melhor desempenho.

Apresentamos algumas aplicações de ferramentas estatísticas que são comumente utilizadas na melhoria da qualidade de processos industriais. Inicialmente, desenvolveu-se procedimentos para testar a competência de laboratórios que participam de programas de ensaios de proficiência. Em situações onde os laboratórios medem várias vezes no mesmo ponto, utilizou-se o modelo de erros de medição, proposto por Jaech [39](1985). Além disso, a inferência sobre os parâmetros de tendência aditiva foi generalizada para a classe de distribuições elípticas. A competência dos laboratórios é avaliada pelo teste da razão de verossimilhança generalizada, do qual, obtemos a distribuição exata para a estatística proposta. Em situações onde os laboratórios medem várias vezes em vários pontos e a variável em análise apresenta variações naturais, utilizou-se o modelo com erro nas variáveis. Diante disso, vamos estender o modelo estrutural definido em Barnett [13] (1969) para o modelo ultra-estrutural com réplicas. Neste caso, vamos avaliar não somente a tendência aditiva, mas também, a tendência multiplicativa, ou seja, avaliar a linearidade das medições. As estimativas dos parâmetros foram obtidas via procedimento do algorítmo EM, com isso, desenvolvemos os teste de Wald, razão de verossimilhança e escore para avaliar a competência dos laboratórios. Nos dois modelos propostos, generalizamos o erro normalizado (En) sugerido pelo Guia 43 [37] para testar a competência dos laboratórios participantes de programas de ensaio de proficiência. Apresentamos também, um procedimento para calcular índices de performance para processos univariados e multivariados. Nestes casos, consideramos que a distribuição dos dados segue uma distribuição Normal assimétrica. Além disso, apresentamos uma análise de simulação onde concluímos que a presença de assimetria nos dados pode causar interpretações erradas sobre o processo, quando a distribuição assumida para os dados é a Normal / We present some applications of statistical tools that are used in the improvement of the quality of industrial processes. Initially, we develop procedures to test the ability of laboratories that participate of programs of proficiency test. In situations where the laboratories measure several times in the same point, we use the model of errors of measurement, considered for Jaech [39](1985). Moreover, the inference on the parameters additive bias was generalized for the class of elliptical distributions. The ability of the laboratories is evaluated by the generalized likelihood ratio test, of which, we get the accurate distribution for the statistics proposal. In situations where the laboratories measure some times in some points and the variable in analysis presents natural variations, uses the model with error in the variable. With this, we go to extend the model structural defined in Barnett [13] (1969) for the ultrastructural model with replicate. In this case, we go to not only evaluate the bias additive, but also, the bias multiplicative, that is, to evaluate the linearity of the measurements. The estimates of the parameters had been gotten by the procedure of the EM algorithm, with this, develop of Wald, likelihood ratio and score test to evaluate the ability of the laboratories. In the two considered models, we generalize the normalized error (En) suggested for Guide 43 [37] to test the ability of the participant laboratories of programs of proficiency test. We also present, a procedure to calculate index of performance for univariate and multivariate processes. In these cases, we consider that the distribution of the data follows a skew Normal distribution. Moreover, we present a simulation analysis where we conclude that the presence of asymmetry in the data can cause interpretations missed on the process, when the distribution assumed for the data is the Normal

Generalized autoregressive moving average (GARMA) models are a class of models that was developed for extending the univariate Gaussian ARMA time series model to a flexible observation-driven model for non-Gaussian time series data. This work presents the GARMA model with discrete distributions and application of resampling techniques to this class of models. We also proposed The Bayesian approach on GARMA models. The TGARMA (Transformed Generalized Autoregressive Moving Average) models was proposed, using the Box-Cox power transformation. Last but not least we proposed the Bayesian approach for the TGARMA (Transformed Generalized Autoregressive Moving Average). / Modelos Autoregressivos e de médias móveis generalizados (GARMA) são uma classe de modelos que foi desenvolvida para extender os conhecidos modelos ARMA com distribuição Gaussiana para um cenário de series temporais não Gaussianas. Este trabalho apresenta os modelos GARMA aplicados a distribuições discretas, e alguns métodos de reamostragem aplicados neste contexto. É proposto neste trabalho uma abordagem Bayesiana para os modelos GARMA. O trabalho da continuidade apresentando os modelos GARMA transformados, utilizando a transformação de Box-Cox. E por último porém não menos importante uma abordagem Bayesiana para os modelos GARMA transformados.

Le schéma de régularisation de Taylor-Lagrange (TLRS) est basé sur la définition des champs en tant que distributions à valeurs d'opérateurs (OPVD). L'expression de ces OPVD implique des fonctions test qui, grâce à leurs propriétés (propriétés d'échelles, super-régularité), permettent d'étendre des distributions singulières à tout l'espace. Ce type de régularisation, que l'on peut qualifier de coupure ultra-douce, est efficace quelque soit le degré de divergence originel et produit des amplitudes finies dépendant d'une échelle intrinsèque sans dimensions. Enfin, ce schéma préserve les symétries du groupe de Poincaré et l'invariance de jauge. Après avoir présenté le formalisme TLRS, celui-ci est appliqué au calcul des corrections radiatives en QED ainsi qu'à celles à la masse du boson de Higgs dans le cadre du modèle standard de la physique des particules. Dans une dernière partie, il est appliqué au modèle de Yukawa dans le cadre de la dynamique sur le front de lumière. Les corrections radiatives et un calcul non-perturbatif d'états liés sont effectués. Ces exemples permettent de vérifier, d'une part, l'applicabilité de ce schéma dans différents cas, et d'autre part, de tester son respect des propriétés de symétrie des théories. / The Taylor-Lagrange regularization scheme (TLRS) is based on the definition of fields as operator valued distributions (OPVD). The expression of these OPVDs implies test functions which, thanks to their properties (scaling properties, super-regularity), allow to extend singular distributions to the whole space. This type of regularization, which could be qualified as an ultra-soft cut-off, is efficient for any order of divergences and produces finite amplitudes depending on an intrinsic dimensionless scale. Finally, this scheme respects the Poincaré group symmetries as well as gauge invariance. After an introduction to the TLRS, it is applied to the calculation of radiative corrections to QED and to the mass of the Higgs boson within the standard model of particle physics. In a last section, it is applied to the Yukawa model using the framework of light front dynamics. Radiative corrections and non-perturbative bound state are calculated. This examples allow to verify, on one hand, the applicability of the TLRS, and on the other hand to test its respect of the symmetry properties of the theories.

Le long des côtes nord-ouest africaines, parmi les plus poissonneuses au monde, les ressources halieutiques sont importantes tant du point de vue socio-économique que culturel. Les petits poissons pélagiques sont suivis au niveau sous régional depuis plus de trois décennies en s’appuyant sur les méthodes hydroacoustiques d’évaluation de stocks. Ces campagnes ont ainsi permis de fournir aux gouvernants une base scientifique en matière de prise de décision à des fins d’aménagement des pêcheries pélagiques. Un défi spécifique est de connaitre leur représentativité par rapport aux stocks ciblés. Nous rassurons sur cette représentativité sur la couverture de Sardinella aurita par ces campagnes. Par contre, pour S. maderensis une partie du stock ne serait pas échantillonnée et la méthode n’est pas adaptée pour Illisha africana et Ethmalosa fimbriata. Nous reportons des déplacements vers le nord dans la répartition de S. aurita et d’autres espèces de pélagique, attribués à la redistribution de l'intensité et de la productivité des upwellings [découpé en 5 sous zones du sud Maroc au sud Sénégal], diminuant ainsi l’abondance de la sardinelle dans la région côtière intertropicale. Pour la première fois en utilisant des séries chronologiques d'observation indépendantes à la méso-échelle, nous montrons un changement vers le nord dans la distribution de S. aurita depuis 1995 et l'attribuons au fort réchauffement de l'océan dans la région. De plus les changements spatiaux observés dans la biomasse de S. aurita au cours des 20 dernières années sont du même ordre de grandeur que ceux enregistrés pour les isothermes de surface. De tels changements auront une incidence sur les considérations politiques dans la gestion de la sécurité alimentaire dans plusieurs pays d'Afrique de l'Ouest. / Along the coast of North-West Africa, among the richest waters over the word, fish supply is important at both socio-economic and cultural levels. In this region, the small pelagic fish are the most abundant marine resources and are monitored using hydroacoustic stock assessment methods since more than three decades. These surveys have allowed providing to local authorities a sound scientific basis for deciding about the management of pelagic fisheries. A specific challenge is to know the representativity of these surveys with respect to the stocks targeted. In this work, we confirm this representativity for Sardinella aurita stock while for S. maderensis, a part of the stock could have not been sampled and for pour Ilisha africana and Ethmalosa fimbriata these surveys are not the best tool. Furthermore, Northward shifts in the distribution of sardinella in the Canary Current Large Marine Ecosystem are recorded in this work and have been attributed to the redistribution of upwelling intensity and productivity, resulting in the abundance of sardinella decreasing in the inter-tropical coastal region. Here for the first time using independent observational time series, we report a robust northward shift in S. aurita since 1995 and attribute it to the strong ocean warming in the region. The observed spatial shifts in biomass in the last 20 years are of the same order of magnitude as those recorded for surface isotherms. Such changes will impact policy considerations in the management of food security in several West African countries and should be considered with respect to “Intended Nationally Determined Contributions”.

This thesis is divided in four chapters. The two first chapters introduce a parametric quantile-based estimation method of univariate heavy-tailed distributions and elliptical distributions, respectively. If one is interested in estimating the tail index without imposing a parametric form for the entire distribution function, but only on the tail behaviour, we propose a multivariate Hill estimator for elliptical distributions in chapter three. In the first three chapters we assume an independent and identically distributed setting, and so as a first step to a dependent setting, using quantiles, we prove in the last chapter the asymptotic normality of marginal sample quantiles for stationary processes under the S-mixing condition.<p><p><p>The first chapter introduces a quantile- and simulation-based estimation method, which we call the Method of Simulated Quantiles, or simply MSQ. Since it is based on quantiles, it is a moment-free approach. And since it is based on simulations, we do not need closed form expressions of any function that represents the probability law of the process. Thus, it is useful in case the probability density functions has no closed form or/and moments do not exist. It is based on a vector of functions of quantiles. The principle consists in matching functions of theoretical quantiles, which depend on the parameters of the assumed probability law, with those of empirical quantiles, which depend on the data. Since the theoretical functions of quantiles may not have a closed form expression, we rely on simulations.<p><p><p>The second chapter deals with the estimation of the parameters of elliptical distributions by means of a multivariate extension of MSQ. In this chapter we propose inference for vast dimensional elliptical distributions. Estimation is based on quantiles, which always exist regardless of the thickness of the tails, and testing is based on the geometry of the elliptical family. The multivariate extension of MSQ faces the difficulty of constructing a function of quantiles that is informative about the covariation parameters. We show that the interquartile range of a projection of pairwise random variables onto the 45 degree line is very informative about the covariation.<p><p><p>The third chapter consists in constructing a multivariate tail index estimator. In the univariate case, the most popular estimator for the tail exponent is the Hill estimator introduced by Bruce Hill in 1975. The aim of this chapter is to propose an estimator of the tail index in a multivariate context; more precisely, in the case of regularly varying elliptical distributions. Since, for univariate random variables, our estimator boils down to the Hill estimator, we name it after Bruce Hill. Our estimator is based on the distance between an elliptical probability contour and the exceedance observations. <p><p><p>Finally, the fourth chapter investigates the asymptotic behaviour of the marginal sample quantiles for p-dimensional stationary processes and we obtain the asymptotic normality of the empirical quantile vector. We assume that the processes are S-mixing, a recently introduced and widely applicable notion of dependence. A remarkable property of S-mixing is the fact that it doesn't require any higher order moment assumptions to be verified. Since we are interested in quantiles and processes that are probably heavy-tailed, this is of particular interest.<p> / Doctorat en Sciences économiques et de gestion / info:eu-repo/semantics/nonPublished

Les séismes sont le résultat de glissements rapides le long de failles actives chargées en contraintes par le mouvement des plaques tectoniques. Il est aujourd'hui établi, au moins pour les grands séismes, que la distribution de ce glissement rapide le long des failles pendant les séismes est hétérogène. Imager la complexité de ces distributions de glissement constitue un enjeu majeur de la sismologie en raison des implications potentielles dans la compréhension de la genèse des séismes et la possibilité associée de mieux anticiper le risque sismique et les tsunamis. Pour améliorer l'imagerie de ces distributions de glissement co-sismique, trois axes peuvent être suivis: augmenter les contraintes sur les modèles en incluant plus d'observations dans les inversions, améliorer la modélisation physique du problème direct et progresser dans le formalisme de résolution du problème inverse. Dans ce travail de thèse, nous explorons ces trois axes à travers l'étude de deux séismes majeurs: les séisme de Tohoku-Oki (Mw 9.0) et de Sumatra-Andaman (Mw 9.1-9.3) survenus en 2011 et 2004, respectivement. / Earthquakes are the results of rapid slip on active faults loaded in stress by the tectonic plates motion. It is now establish - at least for large earthquakes - that the distribution of this rapid slip along the rupturing faults is heterogeneous. Imaging the complexity of such slip distributions is one the main challenges in seismology because of the potential implications on understanding earthquake genesis and the associated possibility to better anticipate devastating shaking and tsunami. To improve the imaging of such co-seismic slip distributions, three axes may be followed: increase the constraints on the source models by including more observations into the inversions, improve the physical modeling of the forward problem and improve the formalism to solve the inverse problem. In this PhD thesis, we explore these three axes by studying two recent major earthquakes: the Tohoku-Oki (Mw 9.0) and Sumatra-Andaman (Mw 9.1-9.3) earthquakes, which occured in 2011 and 2004 respectively.