La structure et le processus de formation des traces latentes dans le poly (allyl diglycol carbonate), PADC, ont été étudiés par spectroscopie FT-IR et par simulation Monte Carlo. La quantité de groupes OH formés est équivalente à la quantité de disparition des groupes éther. L’utilisation de radiations à faible TLE a prouvé que les fonctions carbonyle ne disparaissent que lorsque deux électrons au minimum interagissent avec une seule unité de répétition du polymère. Les résultats obtenus avec des protons de haute énergie permettent de comprendre la différence entre des traces révélables et non-révélables. Sur la base de ces résultats, un nouveau concept physique de Fluence Electronique Radiale autour de la Trace d’un Ion, défini comme la densité d'électrons secondaires qui traversent une surface cylindrique de rayon donné, est proposé pour décrire le seuil de détection du PADC en utilisant le code Geant4-DNA. Les connaissances acquises sont utiles pour trouver des agencements moléculaires appropriés pour de nouveaux détecteurs de sensibilités désirées. / The structure and formation process of latent tracks in poly(allyl diglycol carbonate), PADC, have been examined using the combination of FT-IR spectrometry and a Monte Carlo simulation. The generation amount of OH groups is almost equivalent to the loss amount of ether. An important role of the secondary electron that the carbonyl can be broken only when more than two electrons pass through a single repeat unit is clarified by experiments using low LET radiations. Results of high energy protons lead us to the elucidation of the difference between etchable and un-etchable tracks. Based on these results, a new physical concept of Radial Electron Fluence around Ion Tracks, which is defined as the number density of secondary electron that pass through the cylinder surface with a certain radius is proposed for the detection threshold of PADC using Geant4-DNA. Obtained knowledge is helpful to find appropriate molecule arrangements for new etched track detector with desired sensitivities

Submitted by Daniele Amaral (daniee_ni@hotmail.com) on 2016-09-14T19:37:01Z No. of bitstreams: 1 TeseVAAM.pdf: 2213045 bytes, checksum: 7a27583f2dfac9c11da36b29050f4fe4 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-16T19:48:45Z (GMT) No. of bitstreams: 1 TeseVAAM.pdf: 2213045 bytes, checksum: 7a27583f2dfac9c11da36b29050f4fe4 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-16T19:48:49Z (GMT) No. of bitstreams: 1 TeseVAAM.pdf: 2213045 bytes, checksum: 7a27583f2dfac9c11da36b29050f4fe4 (MD5) / Made available in DSpace on 2016-09-16T19:48:53Z (GMT). No. of bitstreams: 1 TeseVAAM.pdf: 2213045 bytes, checksum: 7a27583f2dfac9c11da36b29050f4fe4 (MD5) Previous issue date: 2015-08-14 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / In this thesis, we construct distribution functions for analysis of lifetimes with the focus in scenes of latent risks inspired in models of the carcinogenesis process. Some properties of these distribution functions are presented and discussed as well as the viability front the models in literature. In some cases the models were describes as long-time model where a part of population presents infinite lifetime, i.e, on the population in study some objects are not susceptible (or immunes) for the event of interest. Some models present complexity in the maximization and was necessary the implementation of routines of maximization that allows us the control some variables that are not offered in routines in tradicional programs. The implemented routine is available in appendix. / Nesta tese apresentamos novas distribuições para o estudo do tempo de vida com foco no cenário de riscos latentes com interpretações das construções dos modelos direcionados com os estudos da carcinogênese. Várias propriedades das distribuições são apresentadas e discutidas bem como a viabilidade destes novos modelos frente a vários modelos na literatura. Alguns dos modelos são estendidos para modelos de longa duração, que são modelos que permitem que os tempos de vida sejam infinitos ou em outras palavras que existem indivíduos não suscetíveis (ou curados) ao evento de interesse. Pela complexidade de vários modelos obtidos foi necessário a implementação de rotinas de maximização para o controle de variáveis que não são oferecidos nas rotinas de maximização em programas tradicionais e esta rotina está disponibilizada em anexo.

O uso de provas elaboradas na forma de itens, em processos de avaliação para classificação, é uma herança histórica dos séculos XVI e XVII, ainda em uso nos dias atuais tanto na educação formal quanto em processos seletivos, a exemplo dos exames vestibulares. Elaboradas para mensurar conhecimentos, traços latentes que não podem ser medidos diretamente, as provas costumam ser corrigidas considerando unicamente o escore obtido pelo sujeito avaliado, sem contemplar informações importantes relacionadas aos itens das mesmas. O presente trabalho teve como objetivos: (i) utilizar a modelagem baseada na teoria da resposta ao item unidimensional - TRI e multidimensional - TRIM para construir escalas do conhecimento para a prova da FUVEST e (ii) classificar os candidatos aos seis cursos de graduação oferecidos pela Escola Superior de Agricultura \"Luiz de Queiroz\", unidade da Universidade de São Paulo, com base na escala construída. A hipótese imbutida no corpo do trabalho admitiu que o uso da TRIM classifica de forma diferente os candidatos que os atuais métodos utilizados pela FUVEST. Foram utilizados os padrões de respostas dos 2326 candidatos submetidos à prova, para que uma análise unidimensional fosse realizada, sob o enfoque da TRI, gerando uma escala de proficiências . Quatro traços latentes foram diagnosticados no processo avaliativo, por meio da modelagem multidimensional da TRIM, gerando uma escala das quatro dimensões. Uma proposta para classificação dos candidatos é apresentada, baseada na média das proficiências individuais ponderada pelas cargas fatoriais diagnosticadas pela modelagem. Análise comparativa entre os critérios de classificação utilizados pela FUVEST e pela TRIM foram realizados, identificando discordância entre os mesmos. O trabalho apresenta propostas de interpretação pedagógica para as escalas unidimensional e multidimensional e indica a TRIM como o critério complementar para classificação dos candidatos, valorizando informações individuais dos itens e, portanto, utilizando uma avaliação classificatória mais abrangente. / The use of elaborate exams in the form of items, in evaluation procedures for classification, is a historical legacy of the 16th and 17th centuries, still in use today both in formal education and in selective cases such as entrance examinations. Designed to measure knowledge, latent trait that can not be measured directly, the exams are usually corrected considering only the score obtained by the subject, without including important information related to the items of it. This study aimed to: (i) use the modeling approach unidimensional and multidimensional item response theory (IRT and MIRT, respectively), to build knowledge scales of the entrance examination FUVEST/2012; (ii) classifing candidates for the 6 undergraduate courses offered by the \"Luiz de Queiroz\" College of Agriculture , unit of the University of São Paulo, based on the scale then. The hypothesis supposes that the use of MIRT ranked candidates differently than current methods used by FUVEST. We used the patterns of responses of 2326 candidates submitted to the test, so that a one-dimensional analysis was performed under the IRT approach, generating a range of proficiencies. Four latent traits were diagnosed in the evaluation process by means of multidimensional modeling MIRT, generating a scale of four dimensions. A proposal for classification of the candidates is presented, based on the weighted average of the individual proficiencies by the factor loadings diagnosed by modeling. Comparative analysis of the classification criteria used by FUVEST and MIRT were performed by identifying discrepancies between them. This work presents the proposals of the pedagogical interpretation for one-dimensional and multidimensional scales and indicates the MIRT as additional criteria for the candidates, to valorize individual information of the items and therefore using a more comprehensive classification review.

As pesquisas científicas na área da Psiquiatria freqüentemente avaliam características subjetivas de indivíduos como, por exemplo, depressão, ansiedade e fobias. Os dados são coletados através de questionários, cujos itens tentam identificar a presença ou ausência de certos sintomas associados à morbidade psiquiátrica de interesse. Alguns desses itens, entretanto, podem provocar constrangimento em parte dos indivíduos respondedores por abordarem características ou comportamentos socialmente questionáveis ou, até, ilegais. Um modelo da teoria de resposta ao item é proposto neste trabalho visando diferenciar a relação entre a probabilidade de presença do sintoma e a gravidade da morbidade de indivíduos constrangidos e não constrangidos. Itens que necessitam dessa diferenciação são chamados \\textbf{itens com comportamento diferencial}. Adicionalmente, o modelo permite assumir que indivíduos constrangidos em responder um item possam vir a mentir em suas respostas, no sentido de omitir a presença de um sintoma. Aplicações do modelo proposto a dados simulados para questionários com 20 itens mostraram que as estimativas dos parâmetros são próximas aos seus verdadeiros valores. A qualidade das estimativas piora com a diminuição da amostra de indivíduos, com o aumento do número de itens com comportamento diferencial e, principalmente, com o aumento do número de itens com comportamento diferencial suscetíveis à mentira. A aplicação do modelo a um conjunto de dados reais, coletados para avaliar depressão em adolescentes, ilustra a diferença do padrão de resposta do item ``crises de choro\" entre homens e mulheres. / Psychiatric scientific research often evaluate subjective characteristics of the individual such as depression, anxiety and phobias. Data are collected through questionnaires with items that try to identify the presence or absence of certain symptoms associated with the psychiatric disease. Some of these items though could make some people embarrassed since they are related to questionable or even illegal social behaviors. The item response theory model proposed within this work envisions to differentiate the relationship between the probability of the symptom presence and the gravity of the disease of embarrassed and non-embarrassed individuals. Items that need this differentiation are called differential item functioning (dif). Additionally, the model has the assumption that individuals embarrassed with one particular item could lie across other answers to omit a possible condition. Applications of the proposed model to simulated data for a 20-item questionnaire have showed that parameter estimates of the proposed model are close to their real values. The estimate accuracy gets worse as the number of individuals decreases, the number of dif increases, and especially as the number of dif susceptible to lying increases. The application of the model to a group of real data, collected to evaluate teenager depression, shows the difference in the probability of \"crying crisis\" presence between men and women.

Submitted by Ubirajara Cruz (ubirajara.cruz@gmail.com) on 2016-09-21T16:14:51Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) dissertacao_daiana_medeiros.pdf: 661783 bytes, checksum: b730e6b69207fcfbd6fb978846974286 (MD5) / Approved for entry into archive by Aline Batista (alinehb.ufpel@gmail.com) on 2016-09-21T16:46:53Z (GMT) No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) dissertacao_daiana_medeiros.pdf: 661783 bytes, checksum: b730e6b69207fcfbd6fb978846974286 (MD5) / Made available in DSpace on 2016-09-21T16:46:53Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) dissertacao_daiana_medeiros.pdf: 661783 bytes, checksum: b730e6b69207fcfbd6fb978846974286 (MD5) Previous issue date: 2014-02-21 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Infecções pelo herpesvírus bovino tipo 1 (BoHV-1) têm sido amplamente detectadas em rebanhos bovinos brasileiros, e são responsáveis por grandes prejuízos econômicos na bovinocultura. Entre as doenças causadas por esse vírus estão a Rinotraqueíte infecciosa bovina (IBR), Vulvovaginite Pustular Infeciosa (IPV) e Balanopostite Pustular Infecciosa (IBP). O impacto econômico dessa enfermidade é devido principalmente a falhas reprodutivas, como retorno ao estro e abortamentos. A espécie bovina é hospedeira natural do BoHV-1, no entanto, estudos sorológicos realizados em diversos países têm sugerido que búfalos (Bubalus bubalis) podem ser suscetíveis ao herpesvírus bovino tipo 1 (BoHV-1), e outros alfaherpesvirus geneticamente relacionados. Apesar de os bovinos se infectarem naturalmente pelo BoHV-1, a distribuição dessa enfermidade na espécie bubalina não tem sido estudada no Brasil, e as informações são limitadas e suportadas apenas por inquéritos sorológicos. Dessa forma, o objetivo desse trabalho foi detectar a presença de DNA viral de BoHV-1 em 202 gânglios trigêmeos de bubalinos coletados em um abatedouro localizado no município de Pelotas através de uma Semi-Nested PCR (SN-PCR) para detecção parcial do gene da glicoproteína D (gD) do envelope do BoHV-1. Além disso, 242 amostras de soro foram testadas pela técnica de soroneutralização (SN) para a detecção de anticorpos neutralizantes frente a diferentes amostras de herpesvírus de ruminantes: BoHV-1 (cepa Los Angeles), BoHV-5 (cepa Riopel - RP), e BuHV (cepa B6). O DNA de BoHV-1 foi detectado em 30,1% dos animais, e os resultados da SN mostraram que 27,6% dos animais apresentaram anticorpos neutralizantes contra, pelo menos, um dos vírus testados. Os resultados demonstram que o BoHV-1 circula em rebanhos bubalinos na região sul do Brasil, e apesar das reatividade cruzada, anticorpos anti-BoHV-5 e anti-BuHV-B6 também foram identificados na população estudada. Este é o primeiro estudo que utiliza testes moleculares para a detecção de infecções latentes por BoHV-1 na espécie bubalina no Brasil. / Infections caused by the bovine herpesvirus type 1 (BoHV-1) have been widely detected in Brazilian cattle, and they are responsible for major economic losses to the cattle industry. Among the diseases caused by this virus there are infectious rhinotracheitis (IBR), Infectious Pustular vulvovaginitis (IPV) and balanoposthitis Infectious Pustular (IBP). The economic impact of this disease is mainly due to reproductive failures, such as return to estrus and abortions. The bovine is a natural host of BoHV-1, however, serologic studies in several countries have suggested that buffaloes (Bubalus bubalis) may be susceptible to bovine herpesvirus type 1 (BoHV-1), and other genetically related alphaherpesvirus. Despite the cattle become infected naturally by BoHV-1, the distribution of this disease in the buffalo species has not been studied in Brazil, and the information is limited and only supported by serological surveys. Thus, the aim of this study was to detect the presence of viral DNA of BHV - 1 in trigeminal ganglia of 202 buffaloes collected in a slaughterhouse located in the city of Pelotas through a semi -nested PCR (SN - PCR) for detection of partial gene glycoprotein D of BoHV-1. In addition, 242 serum samples were obtained and tested by the serum neutralization (SN ) technique for the detection of neutralizing antibodies against different samples of herpesvirus: BHV -1 ( Los Angeles strain ) , BHV -5 (strain Riopel - RP ) and BuHV (strain B6 ). The BHV -1 DNA was detected in 30.1 % of the animals and the results of SN showed that 27.68 % of the animals had neutralizing antibodies against at least one of the tested viruses. This is the first study that uses molecular tests for the detection of latent BoHV - 1 in buffalo species in Brazil.

Le cannabis est la substance illicite la plus consommée à l’adolescence et ce, mondialement. Nous savons qu’un usage problématique de cannabis est susceptible d’entraîner une multitude de conséquences sur les plans physiques, psychologiques, sociaux et économiques. Or, les consommateurs constituent un groupe au profil hétérogène et n’expérimenteront pas tous des problèmes significatifs reliés à cet usage. Il s’avère donc important d’identifier les usagers à risque élevé de conséquences et de spécifier la nature de ces problèmes afin d’intervenir de façon efficace auprès de cette clientèle. Pourtant, bien que la littérature scientifique soulève l’idée que certains types de consommateurs présentent davantage de risques, l’état des connaissances ne révèle que peu d’informations de nature empirique qui permettraient de distinguer des autres usagers les profils d’adolescents davantage à risque. À cet effet, les typologies disponibles à ce jour sont généralement théoriques ou limitées aux études cliniques, font référence à des problèmes particuliers ou se basent sur des indicateurs spécifiques de la consommation tels que la fréquence d’usage. De plus, aucune étude jusqu’à maintenant n’est en mesure de nous pister quant aux types de problèmes associés plus spécifiquement à certains profils de consommation ni de préciser quels profils sont davantage représentés pour des conséquences données. Cette étude est basée sur les classes latentes de consommation découvertes dans l’étude de Fallu, Brière et Janosz (2014) à partir de plusieurs caractéristiques de consommation et comparées sur plusieurs prédicteurs et problèmes associés à la consommation. Son échantillon est constitué de 1618 élèves consommateurs de cannabis en secondaire 4. Cette étude examine quels sont les problèmes attribués et non-attribués les plus fréquents pour chacune des classes. Elle examine également quelles sont les classes les plus représentées pour différents problèmes attribués et non-attribués. Enfin, elle compare les résultats obtenus pour les problèmes attribués et non-attribués. La stratégie analytique employée a consisté à conduire des analyses descriptives, des analyses de Chi carrés ainsi que des analyses de variance univariée, parfois suivies d’une analyse post-hoc. Les résultats ont démontré que la classe d’appartenance peut prédire la survenue des différents problèmes que rencontrent les jeunes consommateurs, que certaines classes sont plus représentées pour certains problèmes et que les adolescents aux profils les plus lourds sont à risque de sous-estimer certains problèmes liés à leur consommation. L’une des principales découvertes concerne les deux classes précoces. Il semble que la classe des consommateurs modérés précoces présente un profil de problèmes davantage intériorisé et la classe des polyconsommateurs lourds précoces, davantage extériorisé. Les implications de ces résultats sont finalement discutées. / Cannabis is the most widely used illicit substance among adolescents worldwide. We know that cannabis misuse is likely to cause a variety of effects at physical, psychological, social and economic levels. However, consumers are a heterogeneous group and not all will experience significant problems associated with such use. It is therefore important to identify high-risk users and to specify the nature of these problems in order to intervene effectively with this clientele. Yet, although the scientific literature raises the idea that certain types of consumers are at higher risk, the current state of knowledge reveals little empirical information that would distinguish adolescents with higher risk profiles. To this end, the typologies available today are generally limited to theoretical or clinical studies, refer to specific problems, or are based on specific consumption indicators such as frequency of use. Furthermore, no study has yet been able to reveal the types of problems associated specifically with certain consumer profiles or to specify which profiles are more prone to various consequences. The present study is based on a latent class consumption model developed from several consumer characteristics found in the study by Fallu, Briere and Janosz (2014). They were compared with several predictors and problems associated with the consumption. The sample comes from within the context of evaluation of the New Approaches New Solutions (NANS) dropout prevention program, and consists of 1618 students who used cannabis in 10th grade. The subjects were annually evaluated from the 7th grade until the 11th. This study examines which attributed and non-attributed problems are most common for each class. It also examines which classes are most frequently assigned to different attributed and non-attributed problems. Finally, it compares the results obtained for the attributed and non-attributed problems. The analytical strategy used was to conduct descriptive analysis, chi-square analysis and univariate analysis of variance (ANOVA), sometimes followed by post-hoc tests when relevant. The results show that the class can predict the occurrence of various problems that young people face, that some classes are more represented for some problems, and that teenagers with heavier profiles are at risk of underestimating certain problems associated with their consumption. One of the main findings regarding the two early classes. It seems that early-moderate use students have a more internalized problems profile and those with early-heavy and polydrug use, a more externalized problems profile. Finally, the implications of these results are discussed.

[pt] Em processos multivariados, existem diversas variáveis observáveis para serem controladas. Pressupõe-se neste trabalho que os descontroles do processo se devem a causas especiais que atuam em fontes de variação independentes, cada uma destas podendo ser representada por uma variável aleatória não observável, ou latente. Alguma alteração na média de uma dessas variáveis ou um aumento na sua dispersão resultam, respectivamente, em deslocamento da média do vetor x de variáveis observáveis ao longo de uma direção atribuível específica, ou aumento da variabilidade do vetor x nessa direção. Propõe-se então controlar a dispersão de tais processos multivariados por gráficos de controle do desvio-padrão dos valores das projeções do vetor de variáveis observadas em direções específicas, associadas a variações nas variáveis latentes não observáveis do processo. Essas direções são denominadas de direções atribuíveis. Foram desenvolvidos, também, gráficos para média da norma quadrática de um vetor resíduo, a fim de permitir a sinalização da ocorrência de novas fontes de variação ainda desconhecidas ao processo, que levem a um aumento da variabilidade do vetor x em direções não contidas no subespaço das direções atribuíveis. O esquema proposto mostrou-se eficaz para o controle estatístico de causas especiais, atuando sobre as fontes de variação do processo, com a vantagem adicional de identificar automaticamente a variável latente afetada. / [en] In multivariate processes, there are several observable variables to be controlled. It is assumed in this work that loss of control is due to special causes acting in independent sources of variation, each of these being represented by an unobservable random variable, or latent. A change in average of these variables or an increase in dispersion results, respectively, in a displacement of the average of the vector x of the observable variables along a specific assignable direction or in an increase of vector x variability in that direction. It is proposed to control the dispersion of such multivariate processes by means of control charts of the vector projections values of observed variables in specific directions, associated with process changes in latent variables, not observable. We call these directions assignable directions. Graphs of average squared norm of a residual vector were developed to enable the signaling of the occurrence of new sources of variation, yet unknown to the process, that lead to increased vector x variability in directions not contained in the assignable directions subspace. The proposed scheme was shown to be an effective tool for statistical control of special causes acting on the process variation sources, with the added benefit of automatically identification of the affected latent variable.

Les études diagnostiques correspondent à l’ensemble des études cliniques qui ont pour objectif l’évaluation d’un nouveau test diagnostique. Dans la démarche d’évaluation, l’étape centrale est l’évaluation de la performance du nouveau test par estimation de sa sensibilité et de sa spécificité. De manière classique, la performance du nouveau test est évaluée par comparaison à un test de référence supposé parfait, appelé un « gold standard » qui permet la connaissance du statut réel de chaque sujet vis-à-vis de la pathologie étudiée. Cependant, dans de très nombreuses situations cliniques, différentes difficultés existent : l’absence de gold standard parfait, l’impossibilité de réalisation du gold standard à tous les sujets, la dépendance des résultats des tests réalisés, la variabilité de la sensibilité et/ou de la spécificité du test en fonction de certaines conditions de réalisation, la multiple réalisation du test dans le temps ou sa multiple interprétation.Une revue méthodologique systématique a été effectuée pour faire l’état des lieux des méthodes d’inférence bayésienne disponibles dans les études diagnostiques et de leur utilisation en pratique. Le focus sur les méthodes bayésiennes a été retenu du fait de leurs avantages théoriques contrastant avec leur relative sous-utilisation dans le domaine médicale. Actuellement, de nombreuses méthodes ont été proposées pour répondre à ces différentes difficultés, avec des développements très complexes en cas de combinaison de plusieurs difficultés dans une même situation. Nous avons ainsi pu établir une cartographie des combinaisons de méthodes disponibles. Cependant leur utilisation en clinique reste encore limitée, même si elle est en augmentation ces dernières années.En pratique, nous avons été confrontés à la problématique du diagnostic de pneumopathie à Pneumocystis jirovecii (PJ) (champignon ubiquitaire opportuniste responsable de mycose profonde chez les patients immunodéprimés). Dans ce projet, nous disposions des résultats de quatre techniques de PCR (Polymerase chain reaction) différentes mais sans gold standard, avec la difficulté supplémentaire de dépendance conditionnelle entre les tests du fait du principe commun à l’origine de ces quatre tests. Deux développements ont été réalisés en parallèle pour répondre à cette problématique : d’une part, un travail sur les méthodes d’élicitation des informations a priori adaptées spécifiquement aux études diagnostiques, et d’autre part, un travail de mise en œuvre d’un modèle statistique adapté à la problématique de quatre tests dépendants en l’absence de gold standard. En l’absence de données informatives dans la littérature, l’élicitation des a priori, étape obligatoire pour l’utilisation des méthodes d’inférence bayésienne, est réalisée par l’interrogation d’experts du domaine. Notre travail a consisté en une adaptation des méthodes existantes, disponibles dans le domaine des essais cliniques, spécifiquement aux études diagnostiques pour obtenir des a priori informatifs. Cette méthode a été appliquée à notre cas des PCR diagnostiques pour PJ. L’estimation des performances diagnostiques des tests en l’absence de gold standard repose de manière efficiente sur les modèles à classes latentes. Trois modèles ont été développés pour le cas de deux tests diagnostiques : un modèle à indépendance conditionnelle, un modèle à dépendance conditionnelle à effets fixes et un modèle à dépendance conditionnelle à effets aléatoires. Nous proposons dans cette thèse une adaptation de ces trois modèles à la situation de quatre tests diagnostiques avec une formulation des paramètres permettant une interprétation clinique des covariances entre les tests dans un souci de transmission des méthodes de la théorie à la pratique. Une application et une comparaison de ces modèles ont été faites pour l’estimation des sensibilités et spécificités des quatre techniques de PCR à PJ en utilisant les a priori informatifs obtenus auprès des experts. / Diagnostic studies include all clinical studies the aim of which is the evaluation of a new diagnostic test. In the evaluation process, the main step is the evaluation of the performance of the new test i.e. its sensitivity and specificity. Usually, the performance of a new test is assessed by comparison to a test of reference which is supposed to be perfect, i.e. a "gold standard", and specifies the actual patient’s status for the disease of interest (“Diseased” or “Not-Diseased” status). However, in many clinical situations, different pitfalls exist such as (i) a gold standard is not available, (ii) the gold standard is not applicable to all patients, (iii) a conditional dependence exists between test results, (iv) the performance of a test is not constant and depends on the conditions of achievement of the test, (v) the tests are repeated in time or by several machines or read by several readers, together with multiple interpretation of the results. A systematic methodological review has been performed to inventory all Bayesian inference methods available in the field of diagnostic studies and their use in practice. The focus on Bayesian methods was based on the theoretical advantages of these methods contrasting with their relative underutilization in the medical field. Finally, several interesting methods have been proposed to address methodological issues of diagnostic studies, with very complex developments when several issues were combined in the same clinical situation. We propose to map the development methods and combinations that have already been done or not. However, their clinical use is still limited, although it has increased in recent years.In practice, we met the problem of the diagnosis of pneumonia due to Pneumocystis jirovecii (PJ). PJ is an ubiquitous opportunistic fungus leading to deep mycosis in immunocompromised patients. In this study, the results of four PCR (polymerase chain reaction) assays were available, but without any gold standard, and the supplementary difficulty of conditional dependence between tests because the four tests were based on the same principle. Two works were performed in parallel to address this issue: on one hand, an adaptation of methods to elicit prior information specifically in diagnostic studies, and on the other hand, the implementation of specific Bayesian statistical models adapted to the context of four-dependent tests in the absence of gold standard. When informative information is not available in the literature, the elicitation of priors, the mandatory first step of a Bayesian inference, is carried out by registering experts’ beliefs in the field. Our work consisted in an adaptation of existing methods, available in clinical trials, specifically for diagnostic studies to obtain informative priors. We then applied this method to our four PJ PCR assays. Estimation of the diagnostic test performance in absence of gold standard is efficiently based on latent class models (LCM). Three LCM were developed for the case of two diagnostic tests: a simple LCM assuming conditional independence between tests, a fixed effects LCM and a random effects LCM providing an adjustment for conditional dependence between tests. We extended these three models to a situation where four diagnostic tests are involved and proposed a formulation that enables an interpretation of between tests covariances in a clinical perspective in order to bind theory to practice. These models were then applied and compared in an estimation study of the sensitivities and specificities of the four PJ PCR assays, by using informative priors obtained from experts.

Depuis les années 70, la littérature économique sur les systèmes financiers a développé une série de concepts théoriques autour des notions de rationnement, de défaillances de marchés et d’imperfections informationnelles, afin de mieux saisir la complexité réelle inhérente au fonctionnement de ces systèmes. Etant donné le caractère vivant, les mutations et la vitesse à laquelle ces mutations s’opèrent dans les systèmes financiers, la question, qui pourtant semblait avoir trouvé un espace consensuel d’analyse et de réflexion autour de ces notions de défaillances de marché, mérite d’être, sinon reposée, du moins actualisée.Actualiser le débat sur les défaillances des marchés financiers implique, symétriquement,d’actualiser le débat sur la place et les modalités d’intervention des politiques publiques dans ceux-Ci. Nous pouvons dès lors poser la problématique générale de notre travail :comment définir, ajuster, calibrer des politiques économiques dans des systèmes financiers hétérogènes et en constante mutation ?Afin d’essayer de répondre à cette problématique, il convient de mieux saisir la complexité des interactions entre le système productif et le système financier, dans les pays en développement. Mieux saisir cette complexité implique de mieux comprendre les mutations des systèmes financiers, de leurs institutions et de leurs acteurs. Cela implique également d’intégrer, en amont, une vision théorique acceptant l’hétérogénéité des acteurs productifs et financiers, et de leurs dynamiques respectives.« Mutations » et « hétérogénéité » sont les deux entrées conceptuelles à travers les quelles nous avons choisi d’analyser les systèmes financiers pour le développement. Cependant, ces deux concepts s’engendrent réciproquement, ce qui présente l’avantage de saisir plus finement la complexité de la problématique mais dont le défi reste de trouver une synthèse appropriée.Les enjeux de développement sous-Jacents à cette problématique seront exposés dans le premier chapitre. Il devrait nous permettre de comprendre en quoi un accès différencié et hétérogène des acteurs économiques aux services financiers a pour conséquence la genèse d’inégalités en termes de dynamiques de développement.Cela suppose, par ailleurs, la prise en compte de l’hétérogénéité des mécanismes de rationnement financier.Dans le deuxième chapitre, nous introduirons la notion de « poches multiples de rationnement » pour traiter des problèmes de contraintes financières à plusieurs niveaux de la structure productive. Nous proposerons un modèle théorique permettant de modéliser cette pluralité de situations de rationnement, que nous chercherons à tester empiriquement par la suite, pour le cas du Maroc, à travers des modèles de frontières stochastiques à classes latentes.La dimension spatiale du rationnement sera un élément traité dans le troisième chapitre et nous permettra de comprendre, à la fois les mutations des systèmes financiers (notamment micro-Financiers) et l’accès inégal des régions boliviennes aux services financiers. Après avoir présenté le concept de Rationnement Spatial du Crédit, nous étudierons, empiriquement, les déterminants de ce phénomène.Cette lecture actualisée et approfondie des mécanismes de rationnement et d’accès inégal aux services financiers devrait permettre de jeter les bases d’un système, amélioré, de calibrage des politiques publiques dans le secteur financier. Ce sont là des éléments que l’on explorera dans les deux derniers chapitres, en étudiant, notamment, les systèmes de gestion et la performance des Banques Nationales de Développement en Amérique Latine. / Sine the 70's, the economic literature on financial systems has developed a series of concepts as "credit rationing", "informational asymmetries" and "market failures" in order to better explain financial systems' real functioning. Giving all transformations lived by financial systems in the last decades, market failures and rationing entries need to be actualized in order to better understand those systems. Actualizing market failures debate means, symmetrically, to actualize the debate about public policies on financial systems. So, the main issue of this theses can be exposed : how to define, design and calibrate economic policies in heterogeneous and constantly evolving financial systems ?Main development issues behind this question are presented in the first chapter. This chapter is important in order to understand the heterogeneity of rationing mechanisms in developing countries' financial systems.In the second chapter, we explore this heterogeneity of rationing mechanisms, analyzing Morocco's financial system. Spatial dimensions of credit rationing are introduced in chapter three in order to understand Micro Finance Institutions' location in Bolivia. This actualized lecture of credit rationing phenoma es useful for sesigning better and calibrated policies in financial systems. In chapter four and five, we analyze modern financial policies, especially through the study of National Development Banks' performance in Latin America.

Les modèles d'équations structurelles à variables latentes permettent de modéliser des relations entre des variables observables et non observables. Les deux paradigmes actuels d'estimation de ces modèles sont les méthodes de moindres carrés partiels sur composantes et l'analyse de la structure de covariance. Dans ce travail, après avoir décrit les deux principales méthodes d'estimation que sont PLS et LISREL, nous proposons une approche d'estimation fondée sur la maximisation par algorithme EM de la vraisemblance globale d'un modèle à facteurs latents et à une équation structurelle. Nous en étudions les performances sur des données simulées et nous montrons, via une application sur des données réelles environnementales, comment construire pratiquement un modèle et en évaluer la qualité. Enfin, nous appliquons l'approche développée dans le contexte d'un essai clinique en cancérologie pour l'étude de données longitudinales de qualité de vie. Nous montrons que par la réduction efficace de la dimension des données, l'approche EM simplifie l'analyse longitudinale de la qualité de vie en évitant les tests multiples. Ainsi, elle contribue à faciliter l'évaluation du bénéfice clinique d'un traitement. / Structural equation models enable the modeling of interactions between observed variables and latent ones. The two leading estimation methods are partial least squares on components and covariance-structure analysis. In this work, we first describe the PLS and LISREL methods and, then, we propose an estimation method using the EM algorithm in order to maximize the likelihood of a structural equation model with latent factors. Through a simulation study, we investigate how fast and accurate the method is, and thanks to an application to real environmental data, we show how one can handly construct a model or evaluate its quality. Finally, in the context of oncology, we apply the EM approach on health-related quality-of-life data. We show that it simplifies the longitudinal analysis of quality-of-life and helps evaluating the clinical benefit of a treatment.