No description available.

Estudamos o problema de obtenção de estimadores/preditores ótimos para combinações lineares de respostas coletadas de uma população finita por meio de amostragem aleatória simples. Nesse contexto, estendemos o modelo misto para populações finitas proposto por Stanek, Singer & Lencina (2004, Journal of Statistical Planning and Inference) para casos em que se incluem erros de medida (endógenos e exógenos) e informação auxiliar. Admitindo que as variâncias são conhecidas, mostramos que os estimadores/preditores propostos têm erro quadrático médio menor dentro da classe dos estimadores lineares não viciados. Por meio de estudos de simulação, comparamos o desempenho desses estimadores/preditores empíricos, i.e., obtidos com a substituição das componentes de variância por estimativas, com aquele de competidores tradicionais. Também, estendemos esses modelos para análise de estudos com estrutura do tipo pré-teste/pós-teste. Também por intermédio de simulação, comparamos o desempenho dos estimadores empíricos com o desempenho do estimador obtido por meio de técnicas clássicas de análise de medidas repetidas e com o desempenho do estimador obtido via análise de covariância por meio de mínimos quadrados, concluindo que os estimadores/ preditores empíricos apresentaram um menor erro quadrático médio e menor vício. Em geral, sugerimos o emprego dos estimadores/preditores empíricos propostos para dados com distribuição assimétrica ou amostras pequenas. / We consider optimal estimation of finite population parameters with data obtained via simple random samples. In this context, we extend a finite population mixed model proposed by Stanek, Singer & Lencina (2004, Journal of Statistical Planning and Inference) by including measurement errors (endogenous or exogenous) and auxiliary information. Assuming that variance components are known, we show that the proposed estimators/predictors have the smallest mean squared error in the class of unbiased estimators. Using simulation studies, we compare the performance of the empirical estimators/predictors obtained by replacing variance components with estimates with the performance of a traditional estimator. We also extend the finite population mixed model to data obtained via pretest-posttest designs. Through simulation studies, we compare the performance of the empirical estimator of the difference in gain between groups with the performance of the usual repeated measures estimator and with the performance of the usual analysis of covariance estimator obtained via ordinary least squares. The empirical estimator has smaller mean squared error and bias than the alternative estimators under consideration. In general, we recommend the use of the proposed estimators/ predictors for either asymmetric response distributions or small samples.

Cette thèse est consacrée à l'inférence de modèles stochastiques d'évolution de l'ADN avec dépendance au contexte, l'étude portant spécifiquement sur la classe de modèles stochastiques RN95+YpR. Cette classe de modèles repose sur un renforcement des taux d'occurrence de certaines substitutions en fonction du contexte local, ce qui introduit des phénomènes de dépendance dans l'évolution des différents sites de la séquence d'ADN. Du fait de cette dépendance, le calcul direct de la vraisemblance des séquences observées met en jeu des matrices de dimensions importantes, et est en général impraticable. Au moyen d'encodages spécifiques à la classe RN95+YpR, nous mettons en évidence de nouvelles structures de dépendance spatiales pour ces modèles, qui sont associées à l'évolution des séquences d'ADN sur toute leur histoire évolutive. Ceci rend notamment possible l'utilisation de méthodes numériques particulaires, développées dans le cadre des modèles de Markov cachés, afin d'obtenir des approximations consistantes de la vraisemblance recherchée. Un autre type d'approximation de la vraisemblance, basé sur des vraisemblances composites, est également introduit. Ces méthodes d'approximation de la vraisemblance sont implémentées au moyen d'un code en C++. Elles sont mises en œuvre sur des données simulées afin d'étudier empiriquement certaines de leurs propriétés, et sur des données génomiques, notamment à des fins de comparaison de modèles d'évolution / This thesis is devoted to the inference of context-dependent evolutionary models of DNA sequences, and is specifically focused on the RN95+YPR class of stochastic models. This class of models is based on the reinforcement of some substitution rates depending on the local context, which introduces dependence phenomena between sites in the evolution of the DNA sequence. Because of these dependencies, the direct computation of the likelihood of the observed sequences involves high-dimensional matrices, and is usually infeasible. Through encodings specific to the RN95+YpR class, we highlight new spatial dependence structures for these models, which are related to the evolution of DNA sequences throughout their evolutionary history. This enables the use of particle filter algorithms, developed in the context of hidden Markov models, in order to obtain consistent approximations of the likelihood. Another type of approximation of the likelihood, based on composite likelihoods, is also introduced. These approximation methods for the likelihood are implemented in a C++ program. They are applied on simulated data to empirically investigate some of their properties, and on genomic data, especially for comparison of evolutionary models

L’estimation de trajectoires de débris spatiaux pendant la rentrée atmosphérique est un défi majeur pour les prochaines années, renforcé par plusieurs projets liés à l'enlèvement de débris établis par plusieurs agences spatiales. Cependant, ce problème s’avère complexe du fait des erreurs de modèle et des difficultés d’initialisation des algorithmes d’estimation induites par une mauvaise connaissance de la dynamique des débris suite à leur désintégration pendant la phase de rentrée atmosphérique. Tout estimateur choisi doit donc être robuste vis-à-vis de ces facteurs. L’estimateur à horizon glissant (MHE) est reconnu dans la littérature pour être robuste vis-à-vis d’erreurs de modèle et de mauvaise initialisation, et les travaux de thèse ont montré qu’il était adapté en termes de performances à la problématique de l’estimation des débris en phase de rentrée. En revanche, il se fonde sur une stratégie d’optimisation qui requiert de fait un temps de calcul important. Pour pallier ce problème, une nouvelle structure d’estimation à horizon glissant a été développée, impliquant un temps de calcul faible nécessaire à l’application envisagée. Cette stratégie, appelée « estimateur à horizon glissant avec pré-estimation (MHE-PE)», prend en compte les erreurs de modèle via un estimateur auxiliaire, plutôt que de chercher à obtenir les estimées du bruit d’état sur l’horizon d’estimation, comme le fait la structure de l’estimateur MHE standard. Un théorème garantissant la stabilité de la dynamique de l’erreur d’estimation du MHE-PE a par ailleurs été proposé. Enfin, les performances de cette structure dans le cadre de l’estimation en trois dimensions des trajectoires de débris pendant la phase de rentrée se sont avérées meilleures que celles observées avec des estimateurs classiques. En particulier, sans dégrader la précision et la convergence de l’estimation, l’estimateur MHE-PE requiert moins de temps de calcul du fait du nombre réduit de paramètres à optimiser. / Space debris tracking during atmospheric re-entries will be a crucial challenge in the coming years, emphasized through many projects on space debris mitigation established by space agencies worldwide. However, this problem appears to be complex, due to model errors and difficulties to properly initialize the estimation algorithms, as a result of unknown dynamics of the debris and their disintegrations during the re-entries. A-to-be used estimator for this problem must be robust against these factors. The Moving Horizon Estimator (MHE) is known in the literature to be robust to model errors and bad initialization, and the PhD work has proved its ability to satisfy performances required by the debris tracking during the re-entries. However, its optimization-based framework induces a large computation time. To overcome this, a new MHE structure which requires smaller computation time than the classical MHE has been developed. This strategy, so-called “Moving Horizon Estimator with Pre-Estimation (MHE-PE)” takes into account model errors by using an auxiliary estimator rather than by searching for estimates of the process noise sequence over the horizon as in the classical strategy. A theorem which guarantees the stability of the dynamics of the estimation errors of the MHE-PE has also been proposed. Finally, performances of this structure in the context of 3D space debris tracking during the re-entries have been shown to be better than those obtained with classical estimators including the MHE. In particular, without degrading accuracy of the estimates and convergence of the estimator, the MHE-PE estimator requires smaller computation time than the MHE thanks to its small number of optimization variables.

Estudamos o problema de obtenção de estimadores/preditores ótimos para combinações lineares de respostas coletadas de uma população finita por meio de amostragem aleatória simples. Nesse contexto, estendemos o modelo misto para populações finitas proposto por Stanek, Singer & Lencina (2004, Journal of Statistical Planning and Inference) para casos em que se incluem erros de medida (endógenos e exógenos) e informação auxiliar. Admitindo que as variâncias são conhecidas, mostramos que os estimadores/preditores propostos têm erro quadrático médio menor dentro da classe dos estimadores lineares não viciados. Por meio de estudos de simulação, comparamos o desempenho desses estimadores/preditores empíricos, i.e., obtidos com a substituição das componentes de variância por estimativas, com aquele de competidores tradicionais. Também, estendemos esses modelos para análise de estudos com estrutura do tipo pré-teste/pós-teste. Também por intermédio de simulação, comparamos o desempenho dos estimadores empíricos com o desempenho do estimador obtido por meio de técnicas clássicas de análise de medidas repetidas e com o desempenho do estimador obtido via análise de covariância por meio de mínimos quadrados, concluindo que os estimadores/ preditores empíricos apresentaram um menor erro quadrático médio e menor vício. Em geral, sugerimos o emprego dos estimadores/preditores empíricos propostos para dados com distribuição assimétrica ou amostras pequenas. / We consider optimal estimation of finite population parameters with data obtained via simple random samples. In this context, we extend a finite population mixed model proposed by Stanek, Singer & Lencina (2004, Journal of Statistical Planning and Inference) by including measurement errors (endogenous or exogenous) and auxiliary information. Assuming that variance components are known, we show that the proposed estimators/predictors have the smallest mean squared error in the class of unbiased estimators. Using simulation studies, we compare the performance of the empirical estimators/predictors obtained by replacing variance components with estimates with the performance of a traditional estimator. We also extend the finite population mixed model to data obtained via pretest-posttest designs. Through simulation studies, we compare the performance of the empirical estimator of the difference in gain between groups with the performance of the usual repeated measures estimator and with the performance of the usual analysis of covariance estimator obtained via ordinary least squares. The empirical estimator has smaller mean squared error and bias than the alternative estimators under consideration. In general, we recommend the use of the proposed estimators/ predictors for either asymmetric response distributions or small samples.

No description available.

Heavy-duty diesel truck engines are routinely idled at standstill to provide cab heating or air conditioning, and in addition to supply electricity to comfort units such as radio and TV. Idling is an inefficient and unfavorable process resulting in increased fuel consumption, increased emissions, shortened engine life, impaired driver rest and health, and elevated noise. Hydrogen-fueled, polymer-electrolyte fuel-cell auxiliary power unit (PEFC-APU) as a silent external power supply, working independently of the main engine, is proposed as viable solution for better fuel economy and abatement of idling emissions. In a diesel PEFC-APU, the hydrogen storage problem is circumvented as hydrogen can be generated onboard from diesel by using a catalytic reformer. In order to make catalytic diesel PEFC-APU systems viable for commercialization research is still needed. Two key areas are the development of reforming catalyst and reformer design, which both are the scope of this thesis. For diesel-reforming catalysts, low loadings of Rh and RhPt alloys have proven to exhibit excellent reforming and hydrogen selectivity properties. For the development of a stable reforming catalyst, more studies have to be conducted in order to find suitable promoters and support materials to optimize and sustain the long-term performance of the Rh catalyst. The next step will be full-scale tests carried out at realistic operating conditions in order to fully comprehend the overall reforming process and to validate promising Rh catalysts. This thesis can be divided into two parts; the first part addresses the development of catalysts in the form of washcoated cordierite monoliths for autothermal reforming (ATR) of diesel. A variety of catalyst compositions were developed containing Rh or RhPt as active metals, CeO2, La2O3, MgO, Y2O3 as promoters and Al2O3, CeO2-ZrO2, SiO2 and TiO2 as support materials. The catalysts were tested in a bench-scale reactor and characterized by using N2-BET, XRD, H2 chemisorption, H2-TPR, O2-TPO, XPS and TEM analyses. The second part addresses the development and testing of full-scale reformers at various realistic operating conditions using promising Rh catalysts. The thesis shows that a variety of Rh on alumina catalysts was successfully tested for ATR of diesel (Papers I-IV). Also, zone-coating, meaning adding two washcoats on specific parts of the monolith, was found to have beneficial effects on the ATR catalyst performance (Paper II). In addition, RhPt supported on CeO2-ZrO2 was found to be one of the most active and promising catalyst candidates for ATR of diesel. The superior performance may be attributed to higher reducibility of RhiOx species and greater dispersion of Rh and Pt on the support (Paper IV). Finally, two full-scale diesel reformers were successfully developed and proven capable of providing high fuel conversion and hydrogen production from commercial diesel over selected Rh catalysts (Papers II-III, V-VI). / QC 20110418

No description available.

Although α-keto carbocations have been demonstrated to be viable intermediates in solvolysis reactions, their applications in synthesis are scarce. These species can be considered to be equivalent to “reversed polarity” enolates and, as such, could be useful for the asymmetric formation of carbon-carbon and carbon-heteroatom bonds. In principle, facial selectivity in additions to α-keto carbocations may be induced using easily removed ester, amide, or imide chiral auxiliaries. Efforts to achieve such diastereoselective SN1 reactions of α-keto carbocations are described herein.

Although α-keto carbocations have been demonstrated to be viable intermediates in solvolysis reactions, their applications in synthesis are scarce. These species can be considered to be equivalent to “reversed polarity” enolates and, as such, could be useful for the asymmetric formation of carbon-carbon and carbon-heteroatom bonds. In principle, facial selectivity in additions to α-keto carbocations may be induced using easily removed ester, amide, or imide chiral auxiliaries. Efforts to achieve such diastereoselective SN1 reactions of α-keto carbocations are described herein.