O presente trabalho tem como principal objetivo, propor e testar métodos alternativos de precificação de contratos de seguro agrícola, baseados em um indicador de produtividade regional. A taxa de prêmio é calculada utilizando a abordagem nãoparamétrica de estimação da densidade da produtividade agrícola, a abordagem paramétrica utilizando as distribuições Normal e Beta e modelos hierárquicos Bayesianos. Na recuperação do processo gerador destes dados, são considerados os efeitos temporal, espacial e espaço-temporal visando a predição e a precificação de um contrato de seguro agrícola regional. Os dois primeiros métodos são aplicados a um conjunto de dados de produtividade municipal do Instituto Brasileiro de Geografia e Estatística (IBGE), no período de 1990 a 2002, para as culturas da soja, milho e trigo, no Estado do Paraná. Na análise empírica do modelo Bayesiano, são utilizados dados de produtividade municipal de milho, no Estado do Paraná, nos anos de 1990 a 2002. A escolha do melhor modelo dentre os modelos não-aninhados ajustados, é baseado no critério da preditiva a posteriori. As metodologias utilizadas nesta pesquisa incorporam melhorias no cálculo atuarial da taxa de prêmio, tendo em vista o pequeno número de observações de produtividade agrícola existentes. Além de propor novas metodologias, estudou-se a viabilidade de implantar um esquema de seguro agrícola regional na região de Castro, no Estado do Paraná, levando em conta a quantificação e redução do risco sistêmico proveniente da aquisição do seguro e da correlação da produtividade individual e regional. Para melhor entendimento dos diversos aspectos do problema, é feito um amplo levantamento histórico e principais tendências do seguro agrícola no Brasil e nos EUA, ressaltando os aspectos legal, institucional e operacional. O estudo mostrou que se o seguro regional de produtividade for oferecido na região de Castro, os produtores se beneficiariam devido à redução do risco proveniente do seguro e também devido ao prêmio relativamente menor do que aquele cobrado pelas seguradoras para os mesmos municípios estudados. / This research analyses alternative methods of pricing agricultural insurance contract based on regional yields. The premium rate is calculated using three different approaches: nonparametric method to estimate the density of the agricultural yield; parametric approach fitting the Normal and Beta distributions; and, hierarchical Bayesian models. The data generating process is recovered considering the temporal, spatial and spatio-temporal aspects to make predictions and pricing for area-yield insurance contract. The data used are county yields, collected by the Brazilian Institute of Geography and Statistics (IBGE), 1990 through 2002. The first two methods were applied to soybean, corn and wheat in the State of Paraná. In the Bayesian model, the empirical analysis limited to corn, in the State of the Paraná, from 1990 through 2002. The choice of the best model among the several non-nested models tested was based on the posterior predictive criteria. The methods proposed in this research intend to improve the actuarial calculation of the premium rate, taking into account the small size of data regarding agricultural yields. Besides proposing different methodologies, a case study of the viability was carried out. The possibility of implementation of an are-yield agricultural insurance was studied in the region of Castro, in the State of the Paraná. This case study considers the quantification and reduction of the systemic risk and also the correlation of the individual and regional yield. To better understand the problem involving the agricultural insurance, a broad historical review of literature was made in Brazil and U.S.A., considering its legal, institutional and operational aspects. The study shows that if a regional yield insurance contract is offered in the Castro region, producers would benefit from exposure to lower risk levels and also a relatively smaller premium rate than the rates charged by insurance companies in the same region.

O presente trabalho tem como principal objetivo, propor e testar métodos alternativos de precificação de contratos de seguro agrícola, baseados em um indicador de produtividade regional. A taxa de prêmio é calculada utilizando a abordagem nãoparamétrica de estimação da densidade da produtividade agrícola, a abordagem paramétrica utilizando as distribuições Normal e Beta e modelos hierárquicos Bayesianos. Na recuperação do processo gerador destes dados, são considerados os efeitos temporal, espacial e espaço-temporal visando a predição e a precificação de um contrato de seguro agrícola regional. Os dois primeiros métodos são aplicados a um conjunto de dados de produtividade municipal do Instituto Brasileiro de Geografia e Estatística (IBGE), no período de 1990 a 2002, para as culturas da soja, milho e trigo, no Estado do Paraná. Na análise empírica do modelo Bayesiano, são utilizados dados de produtividade municipal de milho, no Estado do Paraná, nos anos de 1990 a 2002. A escolha do melhor modelo dentre os modelos não-aninhados ajustados, é baseado no critério da preditiva a posteriori. As metodologias utilizadas nesta pesquisa incorporam melhorias no cálculo atuarial da taxa de prêmio, tendo em vista o pequeno número de observações de produtividade agrícola existentes. Além de propor novas metodologias, estudou-se a viabilidade de implantar um esquema de seguro agrícola regional na região de Castro, no Estado do Paraná, levando em conta a quantificação e redução do risco sistêmico proveniente da aquisição do seguro e da correlação da produtividade individual e regional. Para melhor entendimento dos diversos aspectos do problema, é feito um amplo levantamento histórico e principais tendências do seguro agrícola no Brasil e nos EUA, ressaltando os aspectos legal, institucional e operacional. O estudo mostrou que se o seguro regional de produtividade for oferecido na região de Castro, os produtores se beneficiariam devido à redução do risco proveniente do seguro e também devido ao prêmio relativamente menor do que aquele cobrado pelas seguradoras para os mesmos municípios estudados. / This research analyses alternative methods of pricing agricultural insurance contract based on regional yields. The premium rate is calculated using three different approaches: nonparametric method to estimate the density of the agricultural yield; parametric approach fitting the Normal and Beta distributions; and, hierarchical Bayesian models. The data generating process is recovered considering the temporal, spatial and spatio-temporal aspects to make predictions and pricing for area-yield insurance contract. The data used are county yields, collected by the Brazilian Institute of Geography and Statistics (IBGE), 1990 through 2002. The first two methods were applied to soybean, corn and wheat in the State of Paraná. In the Bayesian model, the empirical analysis limited to corn, in the State of the Paraná, from 1990 through 2002. The choice of the best model among the several non-nested models tested was based on the posterior predictive criteria. The methods proposed in this research intend to improve the actuarial calculation of the premium rate, taking into account the small size of data regarding agricultural yields. Besides proposing different methodologies, a case study of the viability was carried out. The possibility of implementation of an are-yield agricultural insurance was studied in the region of Castro, in the State of the Paraná. This case study considers the quantification and reduction of the systemic risk and also the correlation of the individual and regional yield. To better understand the problem involving the agricultural insurance, a broad historical review of literature was made in Brazil and U.S.A., considering its legal, institutional and operational aspects. The study shows that if a regional yield insurance contract is offered in the Castro region, producers would benefit from exposure to lower risk levels and also a relatively smaller premium rate than the rates charged by insurance companies in the same region.

Les modèles physico-chimiques destinés à représenter la réalité expérimentale peuvent se révéler inadéquats. C'est le cas du piège à oxyde d'azote, utilisé comme support applicatif de notre thèse, qui est un système catalytique traitant les émissions polluantes du moteur Diesel. Les sorties sont des courbes de concentrations des polluants, qui sont des données fonctionnelles, dépendant de concentrations initiales scalaires.L'objectif initial de cette thèse est de proposer des plans d'expériences ayant un sens pour l'utilisateur. Cependant les plans d'expérience s'appuyant sur des modèles, l'essentiel du travail a conduit à proposer une représentation statistique tenant compte des connaissances des experts, et qui permette de construire ce plan.Trois axes de recherches ont été explorés. Nous avons d'abord considéré une modélisation non fonctionnelle avec le recours à la théorie du krigeage. Puis, nous avons pris en compte la dimension fonctionnelle des réponses, avec l'application et l'extension des modèles à coefficients variables. Enfin en repartant du modèle initial, nous avons fait dépendre les paramètres cinétiques des entrées (scalaires) à l'aide d'une représentation non paramétrique.Afin de comparer les méthodes, il a été nécessaire de mener une campagne expérimentale, et nous proposons une démarche de plan exploratoire, basée sur l’entropie maximale. / Physico-chemical models designed to represent experimental reality may prove to be inadequate. This is the case of nitrogen oxide trap, used as an application support of our thesis, which is a catalyst system treating the emissions of the diesel engine. The outputs are the curves of concentrations of pollutants, which are functional data, depending on scalar initial concentrations.The initial objective of this thesis is to propose experiental design that are meaningful to the user. However, the experimental design relying on models, most of the work has led us to propose a statistical representation taking into account the expert knowledge, and allows to build this plan.Three lines of research were explored. We first considered a non-functional modeling with the use of kriging theory. Then, we took into account the functional dimension of the responses, with the application and extension of varying coefficent models. Finally, starting again from the original model, we developped a model depending on the kinetic parameters of the inputs (scalar) using a nonparametric representation.To compare the methods, it was necessary to conduct an experimental campaign, and we propose an exploratory design approach, based on maximum entropy.

Les modèles latents sont très utilisés en pratique, comme en génomique, économétrie, reconnaissance de parole... Comme la modélisation paramétrique des densités d’émission, c’est-à-dire les lois d’une observation sachant l’état latent, peut conduire à de mauvais résultats en pratique, un récent intérêt pour les modèles latents non paramétriques est apparu dans les applications. Or ces modèles ont peu été étudiés en théorie. Dans cette thèse je me suis intéressée aux propriétés asymptotiques des estimateurs (dans le cas fréquentiste) et de la loi a posteriori (dans le cadre Bayésien) dans deux modèles latents particuliers : les modèles de Markov caché et les modèles de mélange. J’ai tout d’abord étudié la concentration de la loi a posteriori dans les modèles non paramétriques de Markov caché. Plus précisément, j’ai étudié la consistance puis la vitesse de concentration de la loi a posteriori. Enfin je me suis intéressée à l’estimation efficace du paramètre de mélange dans les modèles semi paramétriques de mélange. / Latent models have been widely used in diverse fields such as speech recognition, genomics, econometrics. Because parametric modeling of emission distributions, that is the distributions of an observation given the latent state, may lead to poor results in practice, in particular for clustering purposes, recent interest in using non parametric latent models appeared in applications. Yet little thoughts have been given to theory in this framework. During my PhD I have been interested in the asymptotic behaviour of estimators (in the frequentist case) and the posterior distribution (in the Bayesian case) in two particuliar non parametric latent models: hidden Markov models and mixture models. I have first studied the concentration of the posterior distribution in non parametric hidden Markov models. More precisely, I have considered posterior consistency and posterior concentration rates. Finally, I have been interested in efficient estimation of the mixture parameter in semi parametric mixture models.

Dieser Artikel enthält vier Kapitel. Das erste Kapitel ist berechtigt, '''' lokalen Quantil Regression"und seine Zusammenfassung: Quantil Regression ist eine Technik, bedingte Quantil Kurven zu schätzen. Es bietet ein umfassendes Bild über ein Antwort-Kontingent auf erklärenden Variablen. In einem Rahmen flexible Modellierung ist eine besondere Form der bedingten Quantil-Kurve nicht von vornherein festgelegt. Dies motiviert eine lokale parametrische anstatt einer globalen feste Modell passend Ansatz. Eine nichtparametrische Glättung Schätzung der bedingte Quantil Kurve erfordert, zwischen lokalen Krümmung und stochastische auszugleichen Variabilität. In den ersten Essay empfehlen wir eine lokale Modellauswahl Technik, die eine adaptive Schätzung der bedingte bietet Quantil-Regression-Kurve bei jedem Entwurf-Punkt. Theoretische Ergebnisse behaupten, dass das vorgeschlagene adaptive Verfahren als führt gut als Orakel die würde das Risiko der lokalen Abschätzung für die Aufgabenstellung minimieren. Wir veranschaulichen die Leistung der Trolle. / This article includes four chapters. The first chapter is entitled ``Local Quantile Regression", and its summary: Quantile regression is a technique to estimate conditional quantile curves. It provides a comprehensive picture of a response contingent on explanatory variables. In a flexible modeling framework, a specific form of the conditional quantile curve is not a priori fixed. This motivates a local parametric rather than a global fixed model fitting approach. A nonparametric smoothing estimate of the conditional quantile curve requires to balance between local curvature and stochastic variability. In the first essay, we suggest a local model selection technique that provides an adaptive estimate of the conditional quantile regression curve at each design point. Theoretical results claim that the proposed adaptive procedure performs as good as an oracle which would minimize the local estimation risk for the problem at hand. We illustrate the performance of the procedure by an extensive simulation study and consider a couple of applications: to tail dependence analysis for the Hong Kong stock market and to analysis of the distributions of the risk factors of temperature dynamics.

This dissertation is devoted to studying a fast and analytic approximation method, called the variational Bayesian (VB) method, and aims to give insight into its general applicability and usefulness, and explore its applications to various real-world problems. This work has three main foci: 1) The general applicability and properties; 2) Diagnostics for VB approximations; 3) Variational applications. Generally, the variational inference has been developed in the context of the exponential family, which is open to further development. First, it usually consider the cases in the context of the conjugate exponential family. Second, the variational inferences are developed only with respect to natural parameters, which are often not the parameters of immediate interest. Moreover, the full factorization, which assumes all terms to be independent of one another, is the most commonly used scheme in the most of the variational applications. We show that VB inferences can be extended to a more general situation. We propose a special parameterization for a parametric family, and also propose a factorization scheme with a more general dependency structure than is traditional in VB. Based on these new frameworks, we develop a variational formalism, in which VB has a fast implementation, and not be limited to the conjugate exponential setting. We also investigate its local convergence property, the effects of choosing different priors, and the effects of choosing different factorization scheme. The essence of the VB method relies on making simplifying assumptions about the posterior dependence of a problem. By definition, the general posterior dependence structure is distorted. In addition, in the various applications, we observe that the posterior variances are often underestimated. We aim to develop diagnostics test to assess VB approximations, and these methods are expected to be quick and easy to use, and to require no sophisticated tuning expertise. We propose three methods to compute the actual posterior covariance matrix by only using the knowledge obtained from VB approximations: 1) To look at the joint posterior distribution and attempt to find an optimal affine transformation that links the VB and true posteriors; 2) Based on a marginal posterior density approximation to work in specific low dimensional directions to estimate true posterior variances and correlations; 3) Based on a stepwise conditional approach, to construct and solve a set of system of equations that lead to estimates of the true posterior variances and correlations. A key computation in the above methods is to calculate a uni-variate marginal or conditional variance. We propose a novel way, called the VB Adjusted Independent Metropolis-Hastings (VBAIMH) method, to compute these quantities. It uses an independent Metropolis-Hastings (IMH) algorithm with proposal distributions configured by VB approximations. The variance of the target distribution is obtained by monitoring the acceptance rate of the generated chain. One major question associated with the VB method is how well the approximations can work. We particularly study the mean structure approximations, and show how it is possible using VB approximations to approach model selection tasks such as determining the dimensionality of a model, or variable selection. We also consider the variational application in Bayesian nonparametric modeling, especially for the Dirichlet process (DP). The posterior inference for DP has been extensively studied in the context of MCMC methods. This work presents a a full variational solution for DP with non-conjugate settings. Our solution uses a truncated stick-breaking representation. We propose an empirical method to determine the number of distinct components in a finite dimensional DP. The posterior predictive distribution for DP is often not available in a closed form. We show how to use the variational techniques to approximate this quantity. As a concrete application study, we work through the VB method on regime-switching lognormal models and present solutions to quantify both the uncertainty in the parameters and model specification. Through a series numerical comparison studies with likelihood based methods and MCMC methods on the simulated and real data sets, we show that the VB method can recover exactly the model structure, gives the reasonable point estimates, and is very computationally efficient.