Recruitment of pelagic fish populations is believed to be regulated during the planktonic larval stage due to high rates of mortality during the early life stages. Starvation is thought to be one of the main sources of mortality, despite the fact that there is rarely a strong correlation between the feeding success of larval fish and food availability as measured in the field. This lack of relationship may be caused in part by (i) inadequate sampling of larval fish prey and (ii) the use of total zooplankton abundance or biomass as proxies for larval food availability. Many feeding studies rely on measures of average prey abundance which do not adequately capture the variability, or patchiness, of the prey field as experienced by larval fish. Previous studies have shown that larvae may rely on these patches to increase their feeding success. I assess the variability in the availability of larval fish prey over a range of scales and model the small-scale distribution of prey in Conception Bay, Newfoundland. I show that the greatest variability in zooplankton abundance existed at the meter scale, and that larval fish prey were not randomly distributed within the upper mixed layer. This will impact both how well we can model the stochastic nature of larval fish cohorts, as well as how well we can study larval fish feeding from gut content analyses. Expanding on six years of previous lab and field studies on larval Radiated Shanny (Ulvaria subbifurcata) from Conception Bay, Newfoundland, I assess the feeding success, niche breadth (S) and weight-specific feeding rates (SPC, d-1) of the larvae to determine whether there are size-based patterns evident across the years. I found that both the amount of food in the guts and the niche breadth of larvae increased with larval size. There was a shift from low to high SPC with increasing larval size, suggesting that foraging success increases as the larvae grow. My results suggest that efforts should be made to estimate the variability of prey abundance at scales relevant to larval fish foraging rather than using large-scale average abundance estimates, since small-scale prey patchiness likely plays a role in larval fish feeding dynamics. In addition, the characteristics of zooplankton (density, size and behaviour) should be assessed as not all zooplankton are preyed upon equally by all sizes of larval fish. Overall, this thesis demonstrates that indices based on averages fail to account for the variability in the environment and in individual larval fish, which may be confounding the relationship between food availability and larval growth.

Cette thèse étudie le traitement statistique des images TEP. Plus particulièrement, la distribution binomiale négative est proposée pour modéliser l’activité d’une région mono-tissulaire. Cette représentation a l’avantage de pouvoir prendre en compte les variations d’activité biologique (ou hétérogénéité) d’un même tissu. A partir de ces résultats, il est proposé de modéliser la distribution de l’image TEP entière comme un mélange spatialement cohérent de lois binomiales négatives. Des méthodes Bayésiennes sont considérées pour la segmentation d’images TEP et l’estimation conjointe des paramètres du modèle. La cohérence spatiale inhérente aux tissus biologiques est modélisée par un champ aléatoire de Potts-Markov pour représenter la dépendance locale entre les composantes du mélange. Un algorithme original de Monte Carlo par Chaîne de Markov (MCMC) est utilisé, faisant appel aux notions d’échantillonnage dans un espace Riemannien et d’opérateurs proximaux. L’approche proposée est appliquée avec succès à la segmentation de tumeurs en imagerie TEP. Cette méthode est ensuite étendue d’une part en intégrant au processus de segmentation des informations anatomiques acquises par tomodensitométrie (TDM), et d’autre part en traitant une série temporelle d’images correspondant aux différentes phases de respiration. Un modèle de mélange de distributions bivariées binomiale négative - normale est proposé pour représenter les images dynamiques TEP et TDM fusionnées. Un modèle Bayésien hiérarchique a été élaboré comprenant un champ de Potts-Markov à quatre dimensions pour respecter la cohérence spatiale et temporelle des images PET-TDM dynamiques. Le modèle proposé montre une bonne qualité d’ajustement aux données et les résultats de segmentation obtenus sont visuellement en concordance avec les structures anatomiques et permettent la délimitation et le suivi de la tumeur. / This thesis studies statistical image processing of PET images. More specifically, the negative binomial distribution is proposed to model the activity of a single tissue. This representation has the advantage to take into account the variations of biological activity (or heterogeneity) within a single tissue. Based on this, it is proposed to model the data of the entire PET image as a spatially coherent finite mixture of negative binomial distributions. Bayesian methods are considered to jointly perform the segmentation and estimate the model parameters. The inherent spatial coherence of the biological tissue is modeled by a Potts-Markov random field to represent the local dependence between the components of the mixture. An original Markov Chain Monte Carlo (MCMC) algorithm is proposed, based on sampling in a Riemannian space and proximal operators. The proposed approach is successfully applied to the segmentation of tumors in PET imaging. This method is further extended by incorporating anatomical information acquired by computed tomography (CT) and processing a time series of images corresponding to the phases of respiration. A mixture model of bivariate negative binomial - normal distributions is proposed to represent the dynamic PET and CT fused images. A hierarchical Bayesian model was developed including a four dimensional Potts-Markov field to enforce the spatiotemporal coherence of dynamic PET-CT images. The proposed model shows a good fit to the data and the segmentation results obtained are visually consistent with the anatomical structures and allow accurate tumor delineation and tracking.

Made available in DSpace on 2016-06-02T20:06:02Z (GMT). No. of bitstreams: 1 2203.pdf: 1300161 bytes, checksum: 2c1f11d939eab9ab849bb04bf2363a53 (MD5) Previous issue date: 2009-02-10 / Financiadora de Estudos e Projetos / This study considers two discrete distributions based on Bernoulli trials: the Binomial and the Negative Binomial. We explore credibility and confidence intervals to estimate the probability of success of each distribution. The main goal is to analyze their performance coverage probability and average range across the parametric space. We also consider point analysis of bayesian estimators and maximum likelihood estimators, whose interest is to confirm through simulation their consistency, bias and mean square error. In this paper the Objective Bayesian Inference is applied through the noninformative Bayes-Laplace prior, Haldane prior, reference prior and least favorable prior. By analyzing the prior distributions in the minimax decision theory context we verified that the least favorable prior distribution has every other considered prior distributions as particular cases when a quadratic loss function is applied, and matches the Bayes-Laplace prior in considering the quadratic weighed loss function for the Binomial model (which was never found in literature). We used the noninformative Bayes-Laplace prior and Jeffreys prior for the Negative Binomial model. Our findings show through coverage probability, average range of bayesian intervals and point estimation that the Objective Bayesian Inference has good frequentist properties for the probability of success of Binomial and Negative Binomial models. The last stage of this study discusses the presence of correlated proportions in matched-pairs (2 × 2 table) of Bernoulli with the goal of obtaining more information in relation of the considered measures for testing the occurrence of correlated proportions. In this sense the Trinomial model and the partial likelihood function were used from the frequentist and bayesian point of view. The Full Bayesian Significance Test (FBST) was used for real data sets and was shown sensitive to parameterization, however, this study was not possible for the frequentist method since distinct methods are needed to be applied to Trinomial model and the partial likelihood function. / Neste estudo são abordadas duas distribuições discretas baseadas em ensaios de Bernoulli, a Binomial e a Binomial Negativa. São explorados intervalos de credibilidade e confiança para estimação da probabilidade de sucesso de ambas as distribuições. A principal finalidade é analisar nos contextos clássico e bayesiano o desempenho da probabilidade de cobertura e amplitude média gerada pelos intervalos de confiança e intervalos de credibilidade ao longo do espaço paramétrico. Considerou-se também a análise dos estimadores pontuais bayesianos e o estimador de máxima verossimilhança, cujo interesse é confirmar por meio de simulação a consistência e calcular o viés e o erro quadrático médio dos mesmos. A Inferência Bayesiana Objetiva é empregada neste estudo por meio das distribuições a priori não-informativas de Bayes-Laplace, de Haldane, de Jeffreys e menos favorável. Ao analisar as distribuições a priori no contexto de teoria de decisões minimax, a distribuição a priori menos favorável resgata as demais citadas ao empregar a função de perda quadrática e coincide com a distribuição a priori de Bayes-Laplace ao considerar a função de perda quadrática ponderada para o modelo Binomial, o que não foi encontrado até o momento na literatura. Para o modelo Binomial Negativa são consideradas as distribuições a priori não-informativas de Bayes-Laplace e de Jeffreys. Com os estudos desenvolvidos pôde-se observar que a Inferência Bayesiana Objetiva para a probabilidade de sucesso dos modelos Binomial e Binomial Negativa apresentou boas propriedades freqüentistas, analisadas a partir da probabilidade de cobertura e amplitude média dos intervalos bayesianos e por meio das propriedades dos estimadores pontuais. A última etapa do trabalho consiste na análise da ocorrência de proporções correlacionadas em pares de eventos de Bernoulli (tabela 2×2) com a finalidade de determinar um possível ganho de informação em relação as medidas consideradas para testar a ocorrência de proporções correlacionadas. Para tanto fez-se uso do modelo Trinomial e da função de verossimilhança parcial tanto numa abordagem clássica quanto bayesiana. Nos conjuntos de dados analisados observou-se a medida de evidência bayesiana (FBST) como sensível à parametrização, já para os métodos clássicos essa comparação não foi possível, pois métodos distintos precisam ser aplicados para o modelo Trinomial e para a função de verossimilhança parcial.

Metodologias para obtenção do tamanho amostral para estimar a concentração de organismos em água de lastro e verificar normas internacionais são desenvolvidas sob uma abordagem bayesiana. Consideramos os critérios da cobertura média, do tamanho médio e da minimização do custo total sob os modelos Poisson com distribuição a priori gama e binomial negativo com distribuição a priori Pearson Tipo VI. Além disso, consideramos um processo Dirichlet como distribuição a priori no modelo Poisson com o propósito de obter maior flexibilidade e robustez. Para fins de aplicação, implementamos rotinas computacionais usando a linguagem R. / Sample size methodologies for estimating the organism concentration in ballast water and for verifying international standards are developed under a Bayesian approach. We consider the criteria of average coverage, of average length and of total cost minimization under the Poisson model with a gamma prior distribution and the negative binomial model with a Pearson type VI prior distribution. Furthermore, we consider a Dirichlet process as a prior distribution in the Poisson model with the purpose to gain more flexibility and robustness. For practical applications, we implemented computational routines using the R language.

Pas de résumé en français / Malaria is a devastating global health problem that affected 219 million people and caused 660,000 deaths in 2010. Inaccurate estimation of the level of infection may have adverse clinical and therapeutic implications for patients, and for epidemiological endpoint measurements. The level of infection, expressed as the parasite density (PD), is classically defined as the number of asexual parasites relative to a microliter of blood. Microscopy of Giemsa-stained thick blood smears (TBSs) is the gold standard for parasite enumeration. Parasites are counted in a predetermined number of high-power fields (HPFs) or against a fixed number of leukocytes. PD estimation methods usually involve threshold values; either the number of leukocytes counted or the number of HPFs read. Most of these methods assume that (1) the distribution of the thickness of the TBS, and hence the distribution of parasites and leukocytes within the TBS, is homogeneous; and that (2) parasites and leukocytes are evenly distributed in TBSs, and thus can be modeled through a Poisson-distribution. The violation of these assumptions commonly results in overdispersion. Firstly, we studied the statistical properties (mean error, coefficient of variation, false negative rates) of PD estimators of commonly used threshold-based counting techniques and assessed the influence of the thresholds on the cost-effectiveness of these methods. Secondly, we constituted and published the first dataset on parasite and leukocyte counts per HPF. Two sources of overdispersion in data were investigated: latent heterogeneity and spatial dependence. We accounted for unobserved heterogeneity in data by considering more flexible models that allow for overdispersion. Of particular interest were the negative binomial model (NB) and mixture models. The dependent structure in data was modeled with hidden Markov models (HMMs). We found evidence that assumptions (1) and (2) are inconsistent with parasite and leukocyte distributions. The NB-HMM is the closest model to the unknown distribution that generates the data. Finally, we devised a reduced reading procedure of the PD that aims to a better operational optimization and a practical assessing of the heterogeneity in the distribution of parasites and leukocytes in TBSs. A patent application process has been launched and a prototype development of the counter is in process.

The premise of this report is to delve into sports betting and whether favourable opportunities can be found, more specifically focusing on over and under odds for number of goals scored in football games of the Premier League. Using historical data from football matches several models are developed, the characteristics of goals warranting the use of probability based Poisson and Negative Binomial models, as well as Bayesian Poisson regression for goal predictions. Once these models were developed odds was found and compared to bookmakers, the results indicated that all models, to varying degrees, find favourable opportunities and profitable betting strategies can be identified. This suggests that bookmakers do not always price betting products according to their true probabilities likely due to book balancing and informational asymmetries. Furthermore it indicates that there is a presence of inefficiencies in the sports betting market. / Den här rapporten kommer djupdyka i betting och huruvida gynnsamma möjligheter kan hittas. Mer specifikt kommer ett fokus ligga på över/under odds för antalet mål i fotbollsmatcher i engelska Premier League. Genom att använda historisk data från fotbollsmatcher utvecklas flera olika statistiska modeller för att förutspå antalet mål i fotbollsmatcher. Skattning av Poisson och Negativ Binomial fördelningar samt utvecklandet av en Bayesiansk Poisson regressionsmodell motiveras av egenskaperna hos antalet mål i fotbollsmatcher. Med dessa modeller, beräknas odds för flera framtida matcher inom Premier League och dessa jämfördes med odds som ges av bettingbolag. Resultaten indikerar att alla modeller kan, i olika stor utsträckning, hitta gynnsamma möjligheter och lönsamma betting strategier kan identifieras. Detta tyder på att bettingbolag inte alltid sätter sina odds enbart baserat på den faktiska sannolikheten, vilket troligtvis beror på att bolagen balanserar sina böcker samt informationsasymmetrier. Dessutom indikerar resultatet på att det finns faktorer på bettingmarknaden som gör marknaden ineffektiv.

Malaria is a devastating global health problem that affected 219 million people and caused 660,000 deaths in 2010. Inaccurate estimation of the level of infection may have adverse clinical and therapeutic implications for patients, and for epidemiological endpoint measurements. The level of infection, expressed as the parasite density (PD), is classically defined as the number of asexual parasites relative to a microliter of blood. Microscopy of Giemsa-stained thick blood smears (TBSs) is the gold standard for parasite enumeration. Parasites are counted in a predetermined number of high-power fields (HPFs) or against a fixed number of leukocytes. PD estimation methods usually involve threshold values; either the number of leukocytes counted or the number of HPFs read. Most of these methods assume that (1) the distribution of the thickness of the TBS, and hence the distribution of parasites and leukocytes within the TBS, is homogeneous; and that (2) parasites and leukocytes are evenly distributed in TBSs, and thus can be modeled through a Poisson-distribution. The violation of these assumptions commonly results in overdispersion. Firstly, we studied the statistical properties (mean error, coefficient of variation, false negative rates) of PD estimators of commonly used threshold-based counting techniques and assessed the influence of the thresholds on the cost-effectiveness of these methods. Secondly, we constituted and published the first dataset on parasite and leukocyte counts per HPF. Two sources of overdispersion in data were investigated: latent heterogeneity and spatial dependence. We accounted for unobserved heterogeneity in data by considering more flexible models that allow for overdispersion. Of particular interest were the negative binomial model (NB) and mixture models. The dependent structure in data was modeled with hidden Markov models (HMMs). We found evidence that assumptions (1) and (2) are inconsistent with parasite and leukocyte distributions. The NB-HMM is the closest model to the unknown distribution that generates the data. Finally, we devised a reduced reading procedure of the PD that aims to a better operational optimization and a practical assessing of the heterogeneity in the distribution of parasites and leukocytes in TBSs. A patent application process has been launched and a prototype development of the counter is in process.

This thesis is divided into three main parts. In the first, we discuss that, although permutation tests can provide exact control of false positives under the reasonable assumption of exchangeability, there are common examples in which global exchangeability does not hold, such as in experiments with repeated measurements or tests in which subjects are related to each other. To allow permutation inference in such cases, we propose an extension of the well known concept of exchangeability blocks, allowing these to be nested in a hierarchical, multi-level definition. This definition allows permutations that retain the original joint distribution unaltered, thus preserving exchangeability. The null hypothesis is tested using only a subset of all otherwise possible permutations. We do not need to explicitly model the degree of dependence between observations; rather the use of such permutation scheme leaves any dependence intact. The strategy is compatible with heteroscedasticity and can be used with permutations, sign flippings, or both combined. In the second part, we exploit properties of test statistics to obtain accelerations irrespective of generic software or hardware improvements. We compare six different approaches using synthetic and real data, assessing the methods in terms of their error rates, power, agreement with a reference result, and the risk of taking a different decision regarding the rejection of the null hypotheses (known as the resampling risk). In the third part, we investigate and compare the different methods for assessment of cortical volume and area from magnetic resonance images using surface-based methods. Using data from young adults born with very low birth weight and coetaneous controls, we show that instead of volume, the permutation-based non-parametric combination (NPC) of thickness and area is a more sensitive option for studying joint effects on these two quantities, giving equal weight to variation in both, and allowing a better characterisation of biological processes that can affect brain morphology.

Made available in DSpace on 2016-06-02T20:06:08Z (GMT). No. of bitstreams: 1 5190.pdf: 1093269 bytes, checksum: 0d9bf9c7a3855887a0f66859b3a9cc22 (MD5) Previous issue date: 2013-05-06 / Financiadora de Estudos e Projetos / In this work, it was estudied the models INGARCH , GLARMA and GARMA to model count time series data with Poisson and Negative Binomial discrete conditional distributions. The main goal was analyze in classic and bayesian approach, the adequability and goodness of fit of these models, also the contruction of credibility intervals about each parameter. To the Bayesian study, was cosiderated a joint prior distribuition that satisfied the conditions of each model and got a posterior distribution. This aproach presents too some criterion selection like (EBIC), (DIC) and ordenaded predictive conditional density (CPO) for Bayesian cases and (BIC) for classic cases. A simulation study was done to check the maximum likelihood estimator consistency in classic approach and has used criterion selection classic and Bayesian to choose the order of each model. An Analysis has made in a real data set realized as final stage as, these data consist the number of financial transactions in 30 minutes. These results have made in a classical and Bayesian approach , and discribed the data caracteristic. / Nesta dissertação estudou-se os modelos INGARCH, GLARMA e GARMA para modelar séries temporais de dados de contagem com as distribuições condicionais de Poisson e Binomial Negativa. A principal finalidade foi analisar no contexto clássico e bayesiano, a adequabilidade e qualidade de ajuste dos modelos em questão, assim como a construção de intervalos de credibilidade dos parâmetros para cada modelo testado. Para a abordagem Bayesiana foram consideradas priori conjugada, satisfazendo as condições de cada modelo em questão, obtendo assim uma distribuição a posteriori. A abordagem proposta apresenta também o cálculo de critérios de seleção de modelos como o (EBIC), (DIC) e densidade condicional preditiva ordenada (CPO) para o caso Bayesiano e (BIC) para a abordagem clássica. Com um estudo de simulação foi possível verificar a consistência dos estimadores de máxima verossimilhança (clássicos) além disso, foi usado critérios de seleção clássicos e Bayesianos para a seleção da ordem de cada um dos modelos. Uma análise de um conjunto de dados reais foi realizada, sendo uma série do número de transações financeiras realizadas em 30 minutos respectiva os mês de novembro de 2011. Estes resultados apresentam que tanto o estudo clássico, quanto o bayesiano, são capazes de descrever bem o comportamento da série e foram eficientes na escolha da ordem do mesmo.

Nous établissons des conditions nécessaires et suffisantes d’existence et d’infinie divisibilité pour des processus ponctuels alpha-déterminantaux et, lorsque alpha est positif, pour leur intensité sous-jacente (en tant que processus de Cox). Dans le cas où l’espace est fini, ces distributions correspondent à des lois binomiales, négatives binomiales et gamma multidimensionnelles. Nous étudions de façon approfondie ces deux derniers cas avec un noyau non nécessairement symétrique. / We establish necessary and sufficient conditions for the existence and infinite divisibility of alpha-determinantal processes and, when alpha is positive, of their underlying intensity (as Cox process). When the space is finite, these distributions correspond to multidimensional binomial, negative binomial and gamma distributions. We make an in-depth study of these last two cases with a non necessarily symmetric kernel.