There has been an on-going debate about choices of the most suitable model amongst a variety of model specifications and parameterizations. The first dissertation essay investigates whether asymmetric leptokurtic return distributions such as Hansen's (1994) skewed tdistribution combined with GARCH specifications can outperform mixed GARCH-jump models such as Maheu and McCurdy's (2004) GARJI model incorporating the autoregressive conditional jump intensity parameterization in the discrete-time framework. I find that the more parsimonious GJR-HT model is superior to mixed GARCH-jump models. Likelihood-ratio (LR) tests, information criteria such as AIC, SC, and HQ and Value-at-Risk (VaR) analysis confirm that GJR-HT is one of the most suitable model specifications which gives us both better fit to the data and parsimony of parameterization. The benefits of estimating GARCH models using asymmetric leptokurtic distributions are more substantial for highly volatile series such as emerging stock markets, which have a higher degree of non-normality. Furthermore, Hansen's skewed t-distribution also provides us with an excellent risk management tool evidenced by VaR analysis. The second dissertation essay provides a variety of empirical evidences to support redundancy of stochastic volatility for SP500 index returns when stochastic volatility is taken into account with infinite activity pure Lévy jumps models and the importance of stochastic volatility to reduce pricing errors for SP500 index options without regard to jumps specifications. This finding is important because recent studies have shown that stochastic volatility in a continuous-time framework provides an excellent fit for financial asset returns when combined with finite-activity Merton's type compound Poisson jump-diffusion models. The second essay also shows that stochastic volatility with jumps (SVJ) and extended variance-gamma with stochastic volatility (EVGSV) models perform almost equally well for option pricing, which strongly imply that the type of Lévy jumps specifications is not important factors to enhance model performances once stochastic volatility is incorporated. In the second essay, I compute option prices via improved Fast Fourier Transform (FFT) algorithm using characteristic functions to match arbitrary log-strike grids with equal intervals with each moneyness and maturity of actual market option prices.

Os modelos de regressão logística ordinais são usados para descrever a relação entre uma variável resposta categórica ordinal e uma ou mais variáveis explanatórias. Uma vez ajustado o modelo de regressão, se faz necessário verificar a qualidade do ajuste do modelo. As estatísticas qui-quadrado de Pearson e da razão de verossimilhanças não são adequadas para acessar a qualidade do ajuste do modelo de regressão logística ordinal quando variáveis contínuas estão presentes no modelo. Para este caso, foram propostos os testes de Lipsitz, a versão ordinal do teste de Hosmer-Lemeshow e os testes qui-quadrado e razão de verossimilhanças de Pulkistenis-Robinson. Nesta dissertação é feita uma revisão das técnicas de diagnóstico disponíveis para os Modelos logito cumulativo, Modelos logito categorias adjacentes e Modelos logito razão contínua, bem como uma aplicação a fim de investigar a relação entre a perda auditiva, o equilíbrio e aspectos emocionais nos idosos. / Ordinal regression models are used to describe the relationship between an ordered categorical response variable and one or more explanatory variables which could be discrete or continuous. Once the regression model has been fitted, it is necessary to check the goodness-of-fit of the model. The Pearson and likelihood-ratio statistics are not adequate for assessing goodness-of-fit in ordinal logistic regression model with continuous explanatory variables. For this case, the Lipsitz test, the ordinal version of the Hosmer-Lemeshow test and Pulkstenis-Robinson chi-square and likelihood ratio tests were proposed. This dissertation aims to review the diagnostic techniques available for the cumulative logit models, categories adjacent logit models and continuous ratio logistic models. In addition, an application was developed in order to investigate the relationship between hearing loss, balance and emotional aspects in the elderly.

Em experimentos agronômicos são comuns ensaios planejados para estudar determinadas culturas por meio de múltiplas mensurações realizadas na mesma unidade amostral ao longo do tempo, espaço, profundidade entre outros. Essa forma com que as mensurações são coletadas geram conjuntos de dados que são chamados de dados longitudinais. Nesse contexto, é de extrema importância a utilização de metodologias estatísticas que sejam capazes de identificar possíveis padrões de variação e correlação entre as mensurações. A possibilidade de inclusão de efeitos aleatórios e de modelagem das estruturas de covariâncias tornou a metodologia de modelos lineares mistos uma das ferramentas mais apropriadas para a realização desse tipo de análise. Entretanto, apesar de todo o desenvolvimento teórico e computacional, a utilização dessa metodologia em delineamentos mais complexos envolvendo dados longitudinais e tratamentos adicionais, como os utilizados na área de forragicultura, ainda é passível de estudos. Este trabalho envolveu o uso do diagrama de Hasse e da estratégia top-down na construção de modelos lineares mistos no estudo de cortes sucessivos de forragem provenientes de um experimento de adubação com boro em alfafa (Medicago sativa L.) realizado no campo experimental da Embrapa Pecuária Sudeste. Primeiramente, considerou-se uma abordagem qualitativa para todos os fatores de estudo e devido à complexidade do delineamento experimental optou-se pela construção do diagrama de Hasse. A incorporação de efeitos aleatórios e seleção de estruturas de covariâncias para os resíduos foram realizadas com base no teste da razão de verossimilhanças calculado a partir de parâmetros estimados pelo método da máxima verossimilhança restrita e nos critérios de informação de Akaike (AIC), Akaike corrigido (AICc) e bayesiano (BIC). Os efeitos fixos foram testados por meio do teste Wald-F e, devido aos efeitos significativos das fontes de variação associadas ao fator longitudinal, desenvolveu-se um estudo de regressão. A construção do diagrama de Hasse foi fundamental para a compreensão e visualização simbólica do relacionamento de todos os fatores presentes no estudo, permitindo a decomposição das fontes de variação e de seus graus de liberdade, garantindo que todos os testes fossem realizados corretamente. A inclusão de efeito aleatório associado à unidade experimental foi essencial para a modelagem do comportamento de cada unidade e a estrutura de componentes de variância com heterogeneidade, incorporada aos resíduos, foi capaz de modelar eficientemente a heterogeneidade de variâncias presente nos diferentes cortes da cultura da alfafa. A verificação do ajuste foi realizada por meio de gráficos de diagnósticos de resíduos. O estudo de regressão permitiu avaliar a produtividade de matéria seca da parte aérea da planta (kg ha-1) de cortes consecutivos da cultura da alfafa, envolvendo a comparação de adubações com diferentes fontes e doses de boro. Os melhores resultados de produtividade foram observados para a combinação da fonte ulexita com as doses 3, 6 e 9 kg ha-1 de boro. / Assays aimed at studying some crops through multiple measurements performed in the same sample unit along time, space, depth etc. have been frequently adopted in agronomical experiments. This type of measurement originates a dataset named longitudinal data, in which the use of statistical procedures capable of identifying possible standards of variation and correlation among measurements has great importance. The possibility of including random effects and modeling of covariance structures makes the methodology of mixed linear models one of the most appropriate tools to perform this type of analysis. However, despite of all theoretical and computational development, the use of such methodology in more complex designs involving longitudinal data and additional treatments, such as those used in forage crops, still needs to be studied. The present work covered the use of the Hasse diagram and the top-down strategy in the building of mixed linear models for the study of successive cuts from an experiment involving boron fertilization in alfalfa (Medicago sativa L.) carried out in the field area of Embrapa Southeast Livestock. First, we considered a qualitative approach for all study factors and we chose the Hasse diagram building due to the model complexity. The inclusion of random effects and selection of covariance structures for residues were performed based on the likelihood ratio test, calculated based on parameters estimated through the restricted maximum likelihood method, the Akaike\'s Information Criterion (AIC), the Akaike\'s information criterion corrected (AICc) and the Bayesian Information Criterion (BIC). The fixed effects were analyzed through the Wald-F test and we performed a regression study due to the significant effects of the variation sources associated with the longitudinal factor. The Hasse diagram building was essential for understanding and symbolic displaying regarding the relation among all factors present in the study, thus allowing variation sources and their degrees of freedom to be decomposed, assuring that all tests were correctly performed. The inclusion of random effect associated with the sample unit was essential for modeling the behavior of each unity. Furthermore, the structure of variance components with heterogeneity, added to the residues, was capable of modeling efficiently the heterogeneity of variances present in the different cuts of alfalfa plants. The fit was checked by residual diagnostic plots. The regression study allowed us to evaluate the productivity of shoot dry matter (kg ha-1) related to successive cuts of alfalfa plants, involving the comparison of fertilization with different boron sources and doses. We observed the best productivity in the combination of the source ulexite with the doses 3, 6 and 9 kg ha-1 boron.

Em analise de dados que apresentam certo grau de assimetria a suposicao que as observações seguem uma distribuição normal, pode resultar ser uma suposição irreal e a aplicação deste modelo pode ocultar características importantes do modelo verdadeiro. Este tipo de situação deu forca á aplicação de modelo assimétricos, destacando-se entre estes a família de distribuições skew-symmetric, desenvolvida por Azzalini (1985). Neste trabalho nos apresentamos uma segunda proposta para a anàlise de dados com presença importante de assimetria e/ou curtose, comparado com a distribuição normal. Nós apresentamos e estudamos algumas propriedades dos modelos alfa-potência e log-alfa-potência, onde também estudamos o problema de estimação, as matrizes de informação observada e esperada de Fisher e o grau do viés dos estimadores mediante alguns processos de simulação. Nós introduzimos um modelo mais estável que o modelo alfa- potência do qual derivamos o caso bimodal desta distribuição e introduzimos os modelos bimodal simêtrico e assimêtrico alfa-potencia. Posteriormente nós estendemos a distribuição alfa-potência para o caso do modelo Birnbaum-Saunders, estudamos as propriedades deste novo modelo, desenvolvemos estimadores para os parametros e propomos estimadores com viés corrigido. Também introduzimos o modelo de regressão alfa-potência para dados censurados e não censurados e para o modelo de regressão log-linear Birnbaum-Saunders; aqui nós derivamos os estimadores dos parâmetros e estudamos algumas técnicas de validação dos modelos. Por ultimo nós fazemos a extensão multivariada do modelo alfa-potência e estudamos alguns processos de estimação dos parâmetros. Para todos os casos estudados apresentam-se ilustrações com dados já analisados previamente com outras suposições de distribuições. / In data analysis where data present certain degree of asymmetry the assunption of normality can result in an unreal situation and the application of this model can hide important caracteristics of the true model. Situations of this type has given strength to the use of asymmetric models with special emphasis on the skew-symmetric distribution developed by Azzalini (1985). In this work we present an alternative for data analysis in the presence of signi¯cant asymmetry or kurtosis, when compared with the normal distribution, as well as other situations that involve such model. We present and study of the properties of the ®-power and log-®-power distributions, where we also study the estimation problem, the observed and expected information matrices and the degree of bias in estimation using simulation procedures. A °exible model version is proposed for the ®-power distribution, following an extension to a bimodal version. Follows next an extension of the Birnbaum-Saunders distribution using the ®-power distribution, where some properties are studied, estimating approaches are developed as well as corrected bias estimator developed. We also develop censored and uncensored regression for the ®-power model and for the log-linear Birnbaum-Saunders regression models, for which model validation techniques are studied. Finally a multivariate extension of the ®-power model is proposed and some estimation procedures are investigated for the model. All the situations investigated were illustrated with data application using data sets previally analysed with other distributions.

O presente trabalho considera o problema de fazer inferência com acurácia para pequenas amostras, tomando por base a estatística da razão de verossimilhanças em modelos de regressão beta. Estes, por sua vez, são úteis para modelar proporções contínuas que são afetadas por variáveis independentes. Deduzem-se as estatísticas da razão de verossimilhanças ajustadas de Skovgaard (Scandinavian Journal of Statistics 28 (2001) 3-32) nesta classe de modelos. Os termos do ajuste, que têm uma forma simples e compacta, podem ser implementados em um software estatístico. São feitas simulações de Monte Carlo para mostrar que a inferência baseada nas estatísticas ajustadas propostas é mais confiável do que a inferência usual baseada na estatística da razão de verossimilhanças. Aplicam-se os resultados a um conjunto real de dados. / We consider the issue of performing accurate small-sample likelihood-based inference in beta regression models, which are useful for modeling continuous proportions that are affected by independent variables. We derive Skovgaards (Scandinavian Journal of Statistics 28 (2001) 3-32) adjusted likelihood ratio statistics in this class of models. We show that the adjustment terms have simple compact form that can be easily implemented from standard statistical software. We presentMonte Carlo simulations showing that inference based on the adjusted statistics we propose is more reliable than that based on the usual likelihood ratio statistic. A real data example is presented.

Em analise de dados que apresentam certo grau de assimetria a suposicao que as observações seguem uma distribuição normal, pode resultar ser uma suposição irreal e a aplicação deste modelo pode ocultar características importantes do modelo verdadeiro. Este tipo de situação deu forca á aplicação de modelo assimétricos, destacando-se entre estes a família de distribuições skew-symmetric, desenvolvida por Azzalini (1985). Neste trabalho nos apresentamos uma segunda proposta para a anàlise de dados com presença importante de assimetria e/ou curtose, comparado com a distribuição normal. Nós apresentamos e estudamos algumas propriedades dos modelos alfa-potência e log-alfa-potência, onde também estudamos o problema de estimação, as matrizes de informação observada e esperada de Fisher e o grau do viés dos estimadores mediante alguns processos de simulação. Nós introduzimos um modelo mais estável que o modelo alfa- potência do qual derivamos o caso bimodal desta distribuição e introduzimos os modelos bimodal simêtrico e assimêtrico alfa-potencia. Posteriormente nós estendemos a distribuição alfa-potência para o caso do modelo Birnbaum-Saunders, estudamos as propriedades deste novo modelo, desenvolvemos estimadores para os parametros e propomos estimadores com viés corrigido. Também introduzimos o modelo de regressão alfa-potência para dados censurados e não censurados e para o modelo de regressão log-linear Birnbaum-Saunders; aqui nós derivamos os estimadores dos parâmetros e estudamos algumas técnicas de validação dos modelos. Por ultimo nós fazemos a extensão multivariada do modelo alfa-potência e estudamos alguns processos de estimação dos parâmetros. Para todos os casos estudados apresentam-se ilustrações com dados já analisados previamente com outras suposições de distribuições. / In data analysis where data present certain degree of asymmetry the assunption of normality can result in an unreal situation and the application of this model can hide important caracteristics of the true model. Situations of this type has given strength to the use of asymmetric models with special emphasis on the skew-symmetric distribution developed by Azzalini (1985). In this work we present an alternative for data analysis in the presence of signi¯cant asymmetry or kurtosis, when compared with the normal distribution, as well as other situations that involve such model. We present and study of the properties of the ®-power and log-®-power distributions, where we also study the estimation problem, the observed and expected information matrices and the degree of bias in estimation using simulation procedures. A °exible model version is proposed for the ®-power distribution, following an extension to a bimodal version. Follows next an extension of the Birnbaum-Saunders distribution using the ®-power distribution, where some properties are studied, estimating approaches are developed as well as corrected bias estimator developed. We also develop censored and uncensored regression for the ®-power model and for the log-linear Birnbaum-Saunders regression models, for which model validation techniques are studied. Finally a multivariate extension of the ®-power model is proposed and some estimation procedures are investigated for the model. All the situations investigated were illustrated with data application using data sets previally analysed with other distributions.

O presente trabalho considera o problema de fazer inferência com acurácia para pequenas amostras, tomando por base a estatística da razão de verossimilhanças em modelos de regressão beta. Estes, por sua vez, são úteis para modelar proporções contínuas que são afetadas por variáveis independentes. Deduzem-se as estatísticas da razão de verossimilhanças ajustadas de Skovgaard (Scandinavian Journal of Statistics 28 (2001) 3-32) nesta classe de modelos. Os termos do ajuste, que têm uma forma simples e compacta, podem ser implementados em um software estatístico. São feitas simulações de Monte Carlo para mostrar que a inferência baseada nas estatísticas ajustadas propostas é mais confiável do que a inferência usual baseada na estatística da razão de verossimilhanças. Aplicam-se os resultados a um conjunto real de dados. / We consider the issue of performing accurate small-sample likelihood-based inference in beta regression models, which are useful for modeling continuous proportions that are affected by independent variables. We derive Skovgaards (Scandinavian Journal of Statistics 28 (2001) 3-32) adjusted likelihood ratio statistics in this class of models. We show that the adjustment terms have simple compact form that can be easily implemented from standard statistical software. We presentMonte Carlo simulations showing that inference based on the adjusted statistics we propose is more reliable than that based on the usual likelihood ratio statistic. A real data example is presented.

Cette étude est le résultat d’une thèse CIFRE de trois ans entre le bureau d’étude d’Airbus (domaine du contrôle de l’avion) et le laboratoire TéSA à Toulouse. L’objectif principal est de proposer, développer et valider une solution logicielle pour la détection et le diagnostic d’un type spécifique de vibrations des gouvernes de profondeur et direction, appelée oscillation en cycle limite (limit cycle oscillation ou LCO en anglais), basée sur les signaux existants dans les avions civils. LCO est un terme mathématique générique définissant un mode périodique indépendant de conditions initiales et se produisant dans des systèmes non linéaires non conservatifs. Dans cette étude, nous nous intéressons au phénomène de LCO induit par les jeux mécaniques dans les gouvernes d’un avion civil. Les conséquences du LCO sont l’augmentation locale de la charge structurelle, la dégradation des qualités de vol, la réduction de la durée de vie de l’actionneur, la dégradation du confort du poste de pilotage et de la cabine, ainsi que l’augmentation des coûts de maintenance. L’état de l’art en matière de détection et de diagnostic du LCO induit par le jeu mécanique est basé sur la sensibilité du pilote aux vibrations et sur le contrôle périodique du jeu sur les gouvernes. Cette étude propose une solution basée sur les données (issues de la boucle d’asservissement des actionneurs qui agissent sur les gouvernes) pour aider au diagnostic du LCO et à l’isolement du jeu mécanique. L’objectif est d’améliorer encore plus la disponibilité des avions et de réduire les coûts de maintenance en fournissant aux compagnies aériennes un signal de contrôle pour le LCO et les jeux mécaniques. Pour cette raison, deux solutions algorithmiques pour le diagnostic des vibrations et des jeux ont été proposées. Un détecteur en temps réel pour la détection du LCO est tout d’abord proposé basé sur la théorie du rapport de vraisemblance généralisé (generalized likelihood ratio test ou GLRT en anglais). Certaines variantes et simplifications sont également proposées pour satisfaire les contraintes industrielles. Un détecteur de jeu mécanique est introduit basé sur l’identification d’un modèle de Wiener. Des approches paramétrique (estimateur de maximum de vraisemblance) et non paramétrique (régression par noyau) sont explorées, ainsi que certaines variantes des méthodes non paramétriques. En particulier, le problème de l’estimation d’un cycle d’hystérésis (choisi comme la non-linéarité de sortie d’un modèle de Wiener) est abordé. Ainsi, les problèmes avec et sans contraintes sont étudiés. Une analyse théorique, numérique (sur simulateur) et expérimentale (données de vol et laboratoire) est réalisée pour étudier les performances des détecteurs proposés et pour identifier les limitations et la faisabilité industrielle. Les résultats numériques et expérimentaux obtenus confirment que le GLRT proposé (et ses variantes / simplifications) est une méthode très efficace pour le diagnostic du LCO en termes de performance, robustesse et coût calculatoire. D’autre part, l’algorithme de diagnostic des jeux mécaniques est capable de détecter des niveaux de jeu relativement importants, mais il ne fournit pas de résultats cohérents pour des niveaux de jeu relativement faibles. En outre, des types d’entrée spécifiques sont nécessaires pour garantir des résultats répétitifs et cohérents. Des études complémentaires pourraient être menées afin de comparer les résultats de GLRT avec une approche Bayésienne et pour approfondir les possibilités et les limites de la méthode paramétrique proposée pour l’identification du modèle de Wiener. / This research study is the result of a 3 years CIFRE PhD thesis between the Airbus design office(Aircraft Control domain) and TéSA laboratory in Toulouse. The main goal is to propose, developand validate a software solution for the detection and diagnosis of a specific type of elevator andrudder vibration, called limit cycle oscillation (LCO), based on existing signals available in flightcontrol computers on board in-series aircraft. LCO is a generic mathematical term defining aninitial condition-independent periodic mode occurring in nonconservative nonlinear systems. Thisstudy focuses on the LCO phenomenon induced by mechanical freeplays in the control surface ofa civil aircraft. The LCO consequences are local structural load augmentation, flight handlingqualities deterioration, actuator operational life reduction, cockpit and cabin comfort deteriorationand maintenance cost augmentation. The state-of-the-art for freeplay induced LCO detection anddiagnosis is based on the pilot sensitivity to vibration and to periodic freeplay check on the controlsurfaces. This study is thought to propose a data-driven solution to help LCO and freeplaydiagnosis. The goal is to improve even more aircraft availability and reduce the maintenance costsby providing to the airlines a condition monitoring signal for LCO and freeplays. For this reason,two algorithmic solutions for vibration and freeplay diagnosis are investigated in this PhD thesis. Areal time detector for LCO diagnosis is first proposed based on the theory of the generalized likeli hood ratio test (GLRT). Some variants and simplifications are also proposed to be compliantwith the industrial constraints. In a second part of this work, a mechanical freeplay detector isintroduced based on the theory of Wiener model identification. Parametric (maximum likelihoodestimator) and non parametric (kernel regression) approaches are investigated, as well as somevariants to well-known nonparametric methods. In particular, the problem of hysteresis cycleestimation (as the output nonlinearity of a Wiener model) is tackled. Moreover, the constrainedand unconstrained problems are studied. A theoretical, numerical (simulator) and experimental(flight data and laboratory) analysis is carried out to investigate the performance of the proposeddetectors and to identify limitations and industrial feasibility. The obtained numerical andexperimental results confirm that the proposed GLR test (and its variants/simplifications) is a very appealing method for LCO diagnostic in terms of performance, robustness and computationalcost. On the other hand, the proposed freeplay diagnostic algorithm is able to detect relativelylarge freeplay levels, but it does not provide consistent results for relatively small freeplay levels. Moreover, specific input types are needed to guarantee repetitive and consistent results. Further studies should be carried out in order to compare the GLRT results with a Bayesian approach and to investigate more deeply the possibilities and limitations of the proposed parametric method for Wiener model identification.

The focus in this thesis is on the calculations of an empirical null distributionfor likelihood ratio tests testing either separable or double separable covariancematrix structures versus an unstructured covariance matrix. These calculationshave been performed for various dimensions and sample sizes, and are comparedwith the asymptotic χ2-distribution that is commonly used as an approximative distribution. Tests of separable structures are of particular interest in cases when data iscollected such that more than one relation between the components of the observationis suspected. For instance, if there are both a spatial and a temporalaspect, a hypothesis of two covariance matrices, one for each aspect, is reasonable.

The purpose of this investigation was to compare the efficacy of three methods for detecting differential item functioning (DIF). The performance of the crossing simultaneous item bias test (CSIBTEST), the item response theory likelihood ratio test (IRT-LR), and logistic regression (LOGREG) was examined across a range of experimental conditions including different test lengths, sample sizes, DIF and differential test functioning (DTF) magnitudes, and mean differences in the underlying trait distributions of comparison groups, herein referred to as the reference and focal groups. In addition, each procedure was implemented using both an all-other anchor approach, in which the IRT-LR baseline model, CSIBEST matching subtest, and LOGREG trait estimate were based on all test items except for the one under study, and a constant anchor approach, in which the baseline model, matching subtest, and trait estimate were based on a predefined subset of DIF-free items. Response data for the reference and focal groups were generated using known item parameters based on the three-parameter logistic item response theory model (3-PLM). Various types of DIF were simulated by shifting the generating item parameters of select items to achieve desired DIF and DTF magnitudes based on the area between the groups' item response functions. Power, Type I error, and Type III error rates were computed for each experimental condition based on 100 replications and effects analyzed via ANOVA. Results indicated that the procedures varied in efficacy, with LOGREG when implemented using an all-other approach providing the best balance of power and Type I error rate. However, none of the procedures were effective at identifying the type of DIF that was simulated.