A maioria das pesquisas na área de alocação de capacidades entre recursos de uma linha de manufatura propõe as formas balanceadas e em bowl como as mais eficientes para o desempenho de linha como um todo. A maior parte destes estudos é baseada em sistemas simplificados de empurrar a produção, desconsiderando sistemas mais atuais de planejamento e controle da produção (PCP). Por outro lado, estudos referentes à eficiência de sistemas de planejamento e controle da produção não consideram o efeito que critérios distintos de alocação de capacidades podem ter seus desempenhos. Este trabalho tem como objetivo estudar o relacionamento entre as políticas de alocação de capacidades e os sistemas de PCP. O principal critério de desempenho adotado foi o throughput, obtido segundo níveis médios e máximos de estoque em processo em uma linha de produção com cinco recursos. Foram estudados oito tipos de critérios de alocação de capacidades e quatro tipos de sistemas PCP, segundo três níveis de desbalanceamento de cargas, três níveis de coeficiente de variabilidade dos tempos de processamento dos recursos e cinco níveis máximos de estoque em processo. Foi utilizada uma ferramenta de simulação para criar modelos e simular 1386 cenários distintos. Como resultado, percebeu-se uma estreita interdependência entre políticas de alocação e sistemas de PCP. A pesquisa identificou; ainda, que não há um critério de alocação de capacidades nem um sistema de PCP que se mostre melhorem todas as condições testadas, contrariando diversos estudos sobre o tema. / Most of the researches on production capacity allocation among resources, proposes the use of balanced and bowI allocation as the most efficient methods in terms of performance. Such studies were generally based on simplified push production systems, not considering other production pIanning and controI systems (PPC). On the other hand, studies about efficiency of PPC systems don\'t consider the effect of different criteria of capacity allocation on the performance of the PPC systems. The purpose of this research is to investigate how different capacity allocation criteria and different PPC systems interreIate among each other. The major performance criteria used to rank each combination was the resulting throughput, considering several average and maximum levels of work in process (WIP) in a production line with five resources. Eight different types of capacity allocation criteria and four types of PPC systems were studied, with three levels of unbalanced loads, three levels of variability coefficient for processing times and tive maximum WIP levels. A simulation tool was used in order to generate the models and run 1386 different scenarios. As a result, it could be noticed a strong interrelationship between the allocation criteria and the PPC systems. The research also showed, on the contrary of many studies on this subject, that for all the combination tested, none of capacity allocation criteria nor PPC systems stood out on the best option.

J'ai déjà fini le plan de mémoire, et commencé le écriture de mémoire. Je vais envoyer les parties écrits à mon directeur chaque mois. Je pense à finir les parties: Resume, Introduction, Literature: October 2017 Deuxième partie : Octobre 2017 Troisième partie: Novembre 2017 Quatrième partie: Decembre 2017 Cinquième partie: Janvier 2018 Corriger: Fevier 2018 / I have already finished the plan of thesis, and started writing.My project is as below: Summary, Introduction, Literature review: October 2017 Second part : October 2017 Third part: November 2017 Forth part: December 2017 Fifth part: January 2018 Correction and revise: February 2018

Induction machines are the most widely used type of electrical machines because of their robustness, simplicity, and relatively low cost. However, the small airgap in the induction machine makes them more susceptible to Unbalanced Magnetic Pull (UMP). This is because the magnitude of the UMP is a function of the degree of eccentricity, which is the ratio between the length of misalignment and the mean airgap length. The bearing-related failure accounts for approximately 41% of the total failures of induction machines; the percentages of bearing-related failure would be higher for applications in a harsher environment. In this thesis, the UMP caused by rotor eccentricity is investigated, because a small degree of rotor eccentricity is unavoidable due to the manufacturing tolerance and 80% of the mechanical faults could cause rotor eccentricity in electrical machines. When the rotor is not at the centre of the stator, the eccentric rotor causes an uneven airgap around the rotor, in which the magnetic permeance with the higher harmonics content will be created. The magnetomotive force (MMF) produces additional pole-pair ±1 magnetic flux around the airgap. The interaction between each magnetic flux with its pole pair ±1 magnetic flux produces UMP. As only the magnetic flux that crosses the airgap causes UMP, the magnetic flux is categorised into magnetising flux and airgap leakage flux, because both types of flux possess different characteristics at a different rotor slip. As the airgap leakage flux is difficult to calculate analytically, an empirical method is proposed to estimate the UMP caused by the airgap leakage flux. Then, the UMP caused by the magnetising flux can also be estimated by using the empirical method. The parameters for the empirical method can be found by using either the FEA or the experimental results. The damping effect of the magnetising flux in a parallel connected rotor bar is discussed and a damping coefficient is introduced to explain this scenario. The damping coefficient can also be used to calculate the UMP in a steady state analysis. UMP comparisons between the cage rotor and wound rotor induction machines are made. The wound rotor has a much higher UMP because the pole-specific wound rotor could not damp the additional pole pair ±1 magnetic flux. Therefore, a damper winding at the stator slot is also proposed in order to damp the UMP by producing a counteracting flux. In addition, analytical equations have also been derived for different scenarios, such as static eccentricity, dynamic eccentricity, axial-varying eccentricity, and skew rotor bars. Finite Element Analysis (FEA) and experimental work are used to demonstrate the derived analytical equation. Furthermore, the power losses caused by the rotor eccentricity are investigated. Iron losses, copper losses, and frictional loss are discussed and compared with both the analytical equation and the FEA results. In order to reduce the UMP in the induction machines, the two proposed methods are the slip control method and damper windings topology. The slip control method utilises the non-linearity characteristic of the UMP at different rotor slip. To find the optimum operating slip with the lowest UMP, the UMP/Torque ratio is introduced. The characteristics of the UMP/Torque ratio varies with the type and design of the induction machines. However, this method is only applicable when the machine is lightly loaded, because the magnetising flux is limited by the capped terminal voltage and the core saturation of the machine. For the damper winding topology, a circulating current flowing in the damper winding could produce a counteracting flux to damp the UMP. The proposed damper windings configuration is only suitable for the induction machine with an even pole pair number. Finally, comparisons between both UMP reduction methods are made.

A maioria das pesquisas na área de alocação de capacidades entre recursos de uma linha de manufatura propõe as formas balanceadas e em bowl como as mais eficientes para o desempenho de linha como um todo. A maior parte destes estudos é baseada em sistemas simplificados de empurrar a produção, desconsiderando sistemas mais atuais de planejamento e controle da produção (PCP). Por outro lado, estudos referentes à eficiência de sistemas de planejamento e controle da produção não consideram o efeito que critérios distintos de alocação de capacidades podem ter seus desempenhos. Este trabalho tem como objetivo estudar o relacionamento entre as políticas de alocação de capacidades e os sistemas de PCP. O principal critério de desempenho adotado foi o throughput, obtido segundo níveis médios e máximos de estoque em processo em uma linha de produção com cinco recursos. Foram estudados oito tipos de critérios de alocação de capacidades e quatro tipos de sistemas PCP, segundo três níveis de desbalanceamento de cargas, três níveis de coeficiente de variabilidade dos tempos de processamento dos recursos e cinco níveis máximos de estoque em processo. Foi utilizada uma ferramenta de simulação para criar modelos e simular 1386 cenários distintos. Como resultado, percebeu-se uma estreita interdependência entre políticas de alocação e sistemas de PCP. A pesquisa identificou; ainda, que não há um critério de alocação de capacidades nem um sistema de PCP que se mostre melhorem todas as condições testadas, contrariando diversos estudos sobre o tema. / Most of the researches on production capacity allocation among resources, proposes the use of balanced and bowI allocation as the most efficient methods in terms of performance. Such studies were generally based on simplified push production systems, not considering other production pIanning and controI systems (PPC). On the other hand, studies about efficiency of PPC systems don\'t consider the effect of different criteria of capacity allocation on the performance of the PPC systems. The purpose of this research is to investigate how different capacity allocation criteria and different PPC systems interreIate among each other. The major performance criteria used to rank each combination was the resulting throughput, considering several average and maximum levels of work in process (WIP) in a production line with five resources. Eight different types of capacity allocation criteria and four types of PPC systems were studied, with three levels of unbalanced loads, three levels of variability coefficient for processing times and tive maximum WIP levels. A simulation tool was used in order to generate the models and run 1386 different scenarios. As a result, it could be noticed a strong interrelationship between the allocation criteria and the PPC systems. The research also showed, on the contrary of many studies on this subject, that for all the combination tested, none of capacity allocation criteria nor PPC systems stood out on the best option.

As técnicas de mineração de dados, e mais especificamente de aprendizado de máquina, têm se popularizado enormemente nos últimos anos, passando a incorporar os Sistemas de Informação para Apoio à Decisão, Previsão de Eventos e Análise de Dados. Por exemplo, sistemas de apoio à decisão na área médica e ambientes de \\textit{Business Intelligence} fazem uso intensivo dessas técnicas. Algoritmos indutores de árvores de classificação, particularmente os algoritmos TDIDT (Top-Down Induction of Decision Trees), figuram entre as técnicas mais comuns de aprendizado supervisionado. Uma das vantagens desses algoritmos em relação a outros é que, uma vez construída e validada, a árvore tende a ser interpretada com relativa facilidade, sem a necessidade de conhecimento prévio sobre o algoritmo de construção. Todavia, são comuns problemas de classificação em que as frequências relativas das classes variam significativamente. Algoritmos baseados em minimização do erro global de classificação tendem a construir classificadores com baixas taxas de erro de classificação nas classes majoritárias e altas taxas de erro nas classes minoritárias. Esse fenômeno pode ser crítico quando as classes minoritárias representam eventos como a presença de uma doença grave (em um problema de diagnóstico médico) ou a inadimplência em um crédito concedido (em um problema de análise de crédito). Para tratar esse problema, diversos algoritmos TDIDT demandam a calibração de parâmetros {\\em ad-hoc} ou, na ausência de tais parâmetros, a adoção de métodos de balanceamento dos dados. As duas abordagens não apenas introduzem uma maior complexidade no uso das ferramentas de mineração de dados para usuários menos experientes, como também nem sempre estão disponíveis. Neste trabalho, propomos um novo algoritmo indutor de árvores de classificação para problemas com dados desbalanceados. Esse algoritmo, denominado atualmente DDBT (Dynamic Discriminant Bounds Tree), utiliza um critério de partição de nós que, ao invés de se basear em frequências absolutas de classes, compara as proporções das classes nos nós com as proporções do conjunto de treinamento original, buscando formar subconjuntos com maior discriminação de classes em relação ao conjunto de dados original. Para a rotulação de nós terminais, o algoritmo atribui a classe com maior prevalência relativa no nó em relação à prevalência no conjunto original. Essas características fornecem ao algoritmo a flexibilidade para o tratamento de conjuntos de dados com desbalanceamento de classes, resultando em um maior equilíbrio entre as taxas de erro em classificação de objetos entre as classes. / Data mining techniques and, particularly, machine learning methods, have become very popular in recent years. Many decision support information systems and business intelligence tools have incorporated and made intensive use of such techniques. Top-Down Induction of Decision Trees Algorithms (TDIDT) appear among the most popular tools for supervised learning. One of their advantages with respect to other methods is that a decision tree is frequently easy to be interpreted by the domain specialist, precluding the necessity of previous knowledge about the induction algorithms. On the other hand, several typical classification problems involve unbalanced data (heterogeneous class prevalence). In such cases, algorithms based on global error minimization tend to induce classifiers with low error rates over the high prevalence classes, but with high error rates on the low prevalence classes. This phenomenon may be critical when low prevalence classes represent rare or important events, like the presence of a severe disease or the default in a loan. In order to address this problem, several TDIDT algorithms require the calibration of {\\em ad-hoc} parameters, or even data balancing techniques. These approaches usually make data mining tools more complex for less expert users, if they are ever available. In this work, we propose a new TDIDT algorithm for problems involving unbalanced data. This algorithm, currently named DDBT (Dynamic Discriminant Bounds Tree), uses a node partition criterion which is not based on absolute class frequencies, but compares the prevalence of each class in the current node with those in the original training sample. For terminal nodes labeling, the algorithm assigns the class with maximum ration between the relative prevalence in the node and the original prevalence in the training sample. Such characteristics provide more flexibility for the treatment of unbalanced data-sets, yielding a higher equilibrium among the error rates in the classes.

Cada imagem pode ser representada como uma combinação de diversas características, como por exemplo o histograma de intensidades de cor ou propriedades de textura da imagem. Essas características compõem um vetor multidimensional que representa a imagem. É comum esse vetor ser dado como entrada para um método de classificação de padrões que, após aprender por meio de diversos exemplos, pode gerar um modelo de decisão. Estudos sugerem evidências de que a preparação das imagens-- por meio da especificação cuidadosa da aquisição, pré-processamento e segmentação-- pode impactar significativamente a classificação. Além da falta de tratamento das imagens antes da extração de características, o desbalanceamento de classes também se apresenta como um obstáculo para que a classificação seja satisfatória. Imagens possuem características que podem ser exploradas para melhorar a descrição dos objetos de interesse e, portanto, sua classificação. Entre as possibilidades de melhorias estão: a redução do número de intensidades das imagens antes da extração de características ao invés de métodos de quantização no vetor já extraído; e a geração de imagens a partir das originais, de forma a promover o balanceamento de bases de dados cujo número de exemplos de cada classe é desbalanceado. Portanto, a proposta desta dissertação é melhorar a classificação de imagens utilizando métodos de processamento de imagens antes da extração de características. Especificamente, busca analisar a influência do balanceamento de bases de dados e da quantização na classificação. Este estudo analisa ainda a visualização do espaço de características após os métodos de geração artificial de imagens e de interpolação das características extraídas das imagens originais (SMOTE), comparando como espaço original. A ênfase dessa visualização se dá na observação da importância do rebalanceamento das classes. Os resultados obtidos indicam que a quantização simplifica as imagens antes da extração de características e posterior redução de dimensionalidade, produzindo vetores mais compactos; e que o rebalanceamento de classes de imagens através da geração de imagens artificiais pode melhorar a classificação da base de imagens, em relação à classificação original e ao uso de métodos no espaço de características já extraídas. / Each image can be represented by a combination of several features like color frequency and texture properties. Those features compose a multidimensional vector, which represents the original image. Commonly this vector is given as an input to a classification method that can learn from examplesand build a decision model. The literature suggests that image preparation steps like acute acquisition, preprocessing and segmentation can positively impact such classification. Besides that, class unbalancing is also a barrier to achieve good classification accuracy. Some features and methods can be explored to improveobjects\' description, thus their classification. Possible suggestions include: reducing colors number before feature extraction instead of applying quantization methods to raw vectors already extracted; and generating synthetic images from original ones, to balance the number of samples in an uneven data set. We propose to improve image classification using image processing methods before feature extraction. Specifically we want to analyze the influence of both balancing and quantization methods while applied to datasets in a classification routine. This research also analyses the visualization of feature space after the artificial image generation and feature interpolation (SMOTE), against to original space. Such visualization is used because it allows us to know how important is the rebalacing method. The results show that quantization simplifies imagesby producing compacted vectors before feature extraction and dimensionality reduction; and that using artificial generation to rebalance image datasets can improve classification, when compared to the original one and to applying methods on the already extracted feature vectors.

As técnicas de mineração de dados, e mais especificamente de aprendizado de máquina, têm se popularizado enormemente nos últimos anos, passando a incorporar os Sistemas de Informação para Apoio à Decisão, Previsão de Eventos e Análise de Dados. Por exemplo, sistemas de apoio à decisão na área médica e ambientes de \\textit{Business Intelligence} fazem uso intensivo dessas técnicas. Algoritmos indutores de árvores de classificação, particularmente os algoritmos TDIDT (Top-Down Induction of Decision Trees), figuram entre as técnicas mais comuns de aprendizado supervisionado. Uma das vantagens desses algoritmos em relação a outros é que, uma vez construída e validada, a árvore tende a ser interpretada com relativa facilidade, sem a necessidade de conhecimento prévio sobre o algoritmo de construção. Todavia, são comuns problemas de classificação em que as frequências relativas das classes variam significativamente. Algoritmos baseados em minimização do erro global de classificação tendem a construir classificadores com baixas taxas de erro de classificação nas classes majoritárias e altas taxas de erro nas classes minoritárias. Esse fenômeno pode ser crítico quando as classes minoritárias representam eventos como a presença de uma doença grave (em um problema de diagnóstico médico) ou a inadimplência em um crédito concedido (em um problema de análise de crédito). Para tratar esse problema, diversos algoritmos TDIDT demandam a calibração de parâmetros {\\em ad-hoc} ou, na ausência de tais parâmetros, a adoção de métodos de balanceamento dos dados. As duas abordagens não apenas introduzem uma maior complexidade no uso das ferramentas de mineração de dados para usuários menos experientes, como também nem sempre estão disponíveis. Neste trabalho, propomos um novo algoritmo indutor de árvores de classificação para problemas com dados desbalanceados. Esse algoritmo, denominado atualmente DDBT (Dynamic Discriminant Bounds Tree), utiliza um critério de partição de nós que, ao invés de se basear em frequências absolutas de classes, compara as proporções das classes nos nós com as proporções do conjunto de treinamento original, buscando formar subconjuntos com maior discriminação de classes em relação ao conjunto de dados original. Para a rotulação de nós terminais, o algoritmo atribui a classe com maior prevalência relativa no nó em relação à prevalência no conjunto original. Essas características fornecem ao algoritmo a flexibilidade para o tratamento de conjuntos de dados com desbalanceamento de classes, resultando em um maior equilíbrio entre as taxas de erro em classificação de objetos entre as classes. / Data mining techniques and, particularly, machine learning methods, have become very popular in recent years. Many decision support information systems and business intelligence tools have incorporated and made intensive use of such techniques. Top-Down Induction of Decision Trees Algorithms (TDIDT) appear among the most popular tools for supervised learning. One of their advantages with respect to other methods is that a decision tree is frequently easy to be interpreted by the domain specialist, precluding the necessity of previous knowledge about the induction algorithms. On the other hand, several typical classification problems involve unbalanced data (heterogeneous class prevalence). In such cases, algorithms based on global error minimization tend to induce classifiers with low error rates over the high prevalence classes, but with high error rates on the low prevalence classes. This phenomenon may be critical when low prevalence classes represent rare or important events, like the presence of a severe disease or the default in a loan. In order to address this problem, several TDIDT algorithms require the calibration of {\\em ad-hoc} parameters, or even data balancing techniques. These approaches usually make data mining tools more complex for less expert users, if they are ever available. In this work, we propose a new TDIDT algorithm for problems involving unbalanced data. This algorithm, currently named DDBT (Dynamic Discriminant Bounds Tree), uses a node partition criterion which is not based on absolute class frequencies, but compares the prevalence of each class in the current node with those in the original training sample. For terminal nodes labeling, the algorithm assigns the class with maximum ration between the relative prevalence in the node and the original prevalence in the training sample. Such characteristics provide more flexibility for the treatment of unbalanced data-sets, yielding a higher equilibrium among the error rates in the classes.

Dans le cadre du programme de croissance Européen 2020, la commission européenne a mis en place officiellement un chemin à long terme pour une économie à faible émission de carbone, en aspirant une réduction d’au moins 80% des émissions de gaz à effet de serre, d’ici 2050. Répondre à ces exigences ambitieuses, impliquera un changement majeur de paradigme, et notamment en ce qui concerne les infrastructures du réseau électrique. Les percées dans la technologie des semi-conducteurs et les avancées avec les nouvelles topologies d’électronique de puissance et leurs contrôle-commandes, ont contribué à l’impulsion donnée au processus en cours de réaliser un tel SuperGrid. Une percée technologique majeure a eu lieu en 2003, avec le convertisseur modulaire multi-niveaux (MMC ou M2C), présenté par le professeur Marquardt, et qui est actuellement la topologie d’électronique de puissance la plus adaptée pour les stations HVDC. Cependant, cette structure de conversion introduit également un certain nombre de défis relativement complexes tels que les courants “additionnels” qui circulent au sein du convertisseur, entrainant des pertes supplémentaires et un fonctionnement potentiellement instable. Ce projet de thèse vise à concevoir des stratégies de commande “de haut niveau” pour contrôler le MMC adaptées pour les applications à courant continue-haute tension (HVDC), dans des conditions de réseau AC équilibrés et déséquilibrés. La stratégie de commande optimale identifiée est déterminée via une approche pour la conception du type “de haut en bas”, inhérente aux stratégies d’optimisation, où la performance souhaitée du convertisseur MMC donne la stratégie de commande qui lui sera appliquée. Plus précisément, la méthodologie d’optimisation des multiplicateurs de Lagrange est utilisée pour calculer le signal minimal de référence du courant de circulation du MMC dans son repère naturel. / Following Europe’s 2020 growth program, the Energy Roadmap 2050 launched by the European Commission (EC) has officially set a long term path for a low-carbon economy, assuming a reduction of at least 80% of greenhouse gas emissions by the year 2050. Meeting such ambitious requirements will imply a major change in paradigm, including the electricity grid infrastructure as we know it.The breakthroughs in semi-conductor technology and the advances in power electronics topologies and control have added momentum to the on-going process of turning the SuperGrid into a reality. Perhaps the most recent breakthrough occurred in 2003, when Prof. Marquardt introduced the Modular Multilevel Converter (MMC or M2C) which is now the preferred power electronic topology that is starting to be used in VSC-HVDC stations. It does however, introduce a number of rather complex challenges such as “additional” circulating currents within the converter itself, causing extra losses and potentially unstable operation. In addition, the MMC will be required to properly balance the capacitive energy stored within its different arms, while transferring power between the AC and DC grids that it interfaces.The present Thesis project aimed to design adequate “high-level” MMC control strategies suited for HVDC applications, under balanced and unbalanced AC grid conditions. The resulting control strategy is derived with a “top-to-bottom” design approach, inherent to optimization strategies, where the desired performance of the MMC results in the control scheme that will be applied. More precisely, the Lagrange multipliers optimization methodology is used to calculate the minimal MMC circulating current reference signals in phase coordinates, capable of successfully regulating the capacitive arm energies of the converter, while reducing losses and voltage fluctuations, and effectively decoupling any power oscillations that would take place in the AC grid and preventing them from propagating into the DC grid.

Cette thèse aborde le sentiment d'efficacité personnelle (Bandura, 1997/2003) en contexte éducatif plurilingue dans le cadre d'une enquête auprès d'élèves du secondaire en Vallée d'Aoste. Cette région d'Italie est actuellement partagée entre l'idéologie d'un bilinguisme comme addition parfaite de deux langues et la réalité d'un plurilinguisme déséquilibré. L'enquête qualitative repose sur une démarche empirico-déductive qui vise à comprendre la complexité des dynamiques d'un système éducatif plurilingue. Elle s'articule en deux étapes. La première étape est une analyse quantitative du sentiment d'efficacité personnelle en italien et en français des apprenants valdôtains, à l'entrée et à la sortie du secondaire deuxième degré. L'objectif est de comprendre si les apprenants ont une perception équilibrée ou déséquilibrée de leurs compétences en langues et, par conséquent, si la situation valdôtaine présente un bilinguisme équilibré ou un plurilinguisme déséquilibré. Les résultats de cette enquête révèlent aussi que l'épreuve certificative du secondaire en français à l'Esame di Stato ne valorise ni les compétences fortes des apprenants ni les différentes offres formatives des établissements. La deuxième étape de cette enquête qualitative est une recherche-action menée auprès d'élèves de la première et de la deuxième année d'un lycée hôtelier de la Vallée d'Aoste. Le but est de construire un environnement d'enseignement/apprentissage qui vise le développement du sentiment d'efficacité personnelle par le truchement d'activités créatives et par un projet sur ‘apprendre à apprendre'. Cette recherche-action montre que les expériences vicariantes (Bandura, 1988) sont difficiles à mettre en place lorsque le modèle est un autre élève. En revanche, les expériences actives de maîtrise et les états émotionnels sont les deux sources constructrices qui permettent une action concrète en classe développant le sentiment d'efficacité personnelle. Enfin, le guidage cognitif offre l'opportunité de travailler sur les compétences stratégiques afin que l'élève puisse contrôler et réguler son apprentissage. / This PhD thesis focuses on the self-efficacy (Bandura, 1997) in a multilingual educational context through a survey aimed at students of high school in Aosta Valley. This Italian region is currently shared between the ideology of a bilingualism as a perfect addition of two languages and the reality of an unbalanced multilingualism. The qualitative survey bases on an empirical-deductive approach which aims at the understanding of the complex dynamics of a multilingual educational system. The survey articulates in two stages. The first stage is a quantitative analysis of self-efficacy in both the Italian and French languages of the Aostian learners, in the first and the final years of high school. The objective is to understand if the learners have a well-balanced or unbalanced perception of their competences in languages and, consequently, if the Aosta Valley's situation presents a well-balanced bilingualism or an unbalanced multilingualism. The results of this survey also reveal that the final test of high school in French in the Esame di Stato does not value either the strong competences of the learners neither the various formative offers of the institutes. The second stage of this qualitative survey is an action research study undertaken with students of first and second year in a Hotel management school in Aosta Valley. The purpose is to build an environment of education/learning which aims at the development of self-efficacy through creative activities and through a project about ‘learn to learn'. This action research study shows that the vicarious experiences (Bandura, 1988) are difficult to set up when the model is a different student. On the other hand, the mastery of experiences and emotional states are both sources builders who allow a concrete action in class to increase self-efficacy. Finally, the cognitive guide offers the opportunity to work on the strategic competences in order that students could control and regulate their learning.

Submitted by Andressa Souza Campos Monteiro de Castro (dessascmc@gmail.com) on 2015-04-29T19:10:59Z No. of bitstreams: 1 Dissertacao_final.pdf: 676467 bytes, checksum: fdc9136b5dfb8c962d18e88e3f041564 (MD5) / Approved for entry into archive by BRUNA BARROS (bruna.barros@fgv.br) on 2015-04-30T14:49:43Z (GMT) No. of bitstreams: 1 Dissertacao_final.pdf: 676467 bytes, checksum: fdc9136b5dfb8c962d18e88e3f041564 (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2015-05-04T12:47:02Z (GMT) No. of bitstreams: 1 Dissertacao_final.pdf: 676467 bytes, checksum: fdc9136b5dfb8c962d18e88e3f041564 (MD5) / Made available in DSpace on 2015-05-04T12:47:13Z (GMT). No. of bitstreams: 1 Dissertacao_final.pdf: 676467 bytes, checksum: fdc9136b5dfb8c962d18e88e3f041564 (MD5) Previous issue date: 2015-03-20 / This paper investigates the role of consumption-wealth ratio on predicting future stock returns through a panel approach. We follow the theoretical framework proposed by Lettau and Ludvigson (2001), in which a model derived from a nonlinear consumer’s budget constraint is used to settle the link between consumption-wealth ratio and stock returns. Using G7’s quarterly aggregate and financial data ranging from the first quarter of 1981 to the first quarter of 2014, we set an unbalanced panel that we use for both estimating the parameters of the cointegrating residual from the shared trend among consumption, asset wealth and labor income, cay, and performing in and out-of-sample forecasting regressions. Due to the panel structure, we propose different methodologies of estimating cay and making forecasts from the one applied by Lettau and Ludvigson (2001). The results indicate that cay is in fact a strong and robust predictor of future stock return at intermediate and long horizons, but presents a poor performance on predicting one or two-quarter-ahead stock returns.