We determine the dynamical structure factors of two gapped correlated electron systems, namely the Ising model in a strong transverse field and the two-leg spin-1/2 Heisenberg ladder in the limit of strong rung coupling. We consider the low-temperature limit, employing a variety of analytical and numerical techniques. The coherent modes of single-particle excitations, which are delta functions at zero temperature, are shown to broaden asymmetrically in energy with increasing temperature. Firstly, we apply a low-temperature “resummation” inspired by the Dyson equation to a linked-cluster expansion of the two-leg Heisenberg ladder. We include matrix elements to second order in the interaction between states containing up to two particles. A low-frequency response similar to the “Villain mode” is also observed. Next, we apply a cumulant expansion technique to the transverse field Ising model. We resolve the issue of negative spectral weight caused by double pole in the leading self-energy diagram by including a resummation of terms obtained from the six-point function, demonstrating that the perturbation series in 2n-spin correlation functions can be extended to higher orders. The result generalises to higher dimensions and the analytic calculation is compared to a numerical Pade approximant. We outline the extension of this method to the strong coupling ladder. Finally, we compare the previous results to numerical data obtained by full diagonalisation of finite chains and numerical evaluation of the Pfaffian, a method specific to the transverse field Ising chain. The latter method is used for a phenomenological study of the asymmetric broadening as well as an evaluation of fitting functions for the broadened lineshapes.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Claims reserving methods usually use data aggregated into development triangles, therefore a lot of information that insurance companies possess remains unused. This thesis shows a triangle-free approach using granular information from a claim by claim database. A statistical model for claims development which can further be used for estimation of reserves is built. The statistical model consists of a counting process that drives claims occurrence, distribution of reporting delay and distribution of claims severity. Several suitable distributions are presented, as well as methods for obtaining their parameters from data. Theoretical apparatus is used for real data. The thesis also pursues comparison of the IBNR reserve estimation using the triangle free approach and distribution free Chain ladder method for real data as well as for simulated data sets. For the data used in this thesis the complexity and data requirements of the triangle free approach are in favor of more preciseness and versatility. Powered by TCPDF (www.tcpdf.org)

Syftet med studien var att undersöka hur stor andel barn mellan 10–14 år som tydligt fått komma till tals samt hur deras utsagor uttrycks i de skriftliga domsluten vid överklagande av LVU till kammarrätten. En kategorisering av alla domslut från utvald kammarrätt gjordes för att sedan välja ut åtta av domsluten att granska kvalitativt utifrån Harts delaktighetsstege och barndomssociologi. Resultatet visade att barn i mindre än hälften av domsluten tydligt fått komma till tals och att barnets åsikt uttrycks i olika stor utsträckning i de skriftliga domsluten, dock framkommer inte barnets åsikter i avsnittet skäl för beslut i domslutet. Barn anses ofta i förhållande till vuxna inte vara tillräckligt mogna vilket kan medför att de inte får vara aktiva aktörer och att deras handlingsutrymme begränsas. Vuxna som arbetar med barn måste vara medvetna om den komplexitet som råder mellan barnets rätt till delaktighet och barnets bästa. / The aim of this study was to examine the proportion of children between ages 10-14 who clearly got to express themselves and how their views were expressed in written documents when appealing LVU to kammarrätten. A categorization of all judgments from a court were made, the eight were selected to examine qualitatively supported by Hart's ladder of participation and childhood sociology. The result showed that children in less than half of the judgements clearly got to express themselves and that the child's words were noted to varying degrees, however, in the section reasons for decision the child’s views were not included. Children are often not considered sufficiently mature in relation to adults, which means that they may not be active actors and that their room for maneuver is restricted. Adults working with children must be aware of the complexity between the child's right to participation and the child’s best interests.

Nesta tese desenvolvemos um método numérico para diagonalizar o Hamiltoniano de Heisenberg iterativamente. O método consiste basicamente em diagonalizar cadeias de spins, cada vez maiores, em que cada passo da diagonalização corresponde à adição de um novo spin à cadeia. A base de vetores para calcular o Hamiltoniano de uma cadeia de N spins, HN, é construída por meio do produto direto dos autovetores do Hamiltoniano Hn-1 da rede diagonalizada no passo anterior, pelos autoestados correspondentes ao N-ésimo spin adicionado. Além de usar a comutação do Hamiltoniano com a componente azimutal do spin total, Sz, prática comum em outros métodos, usufruímos da conservação com o quadrado do spin total, S2. Para uma classe específica de redes também implementamos a simetria de reflexão. Obtemos o espectro completo de energia de cadeias de spins 1/2 com até 20 sítios, para as quais mostramos resultados da dependência com a temperatura da susceptibilidade magnética e do calor específico, para redes com impurezas tipo spin substitucionais, com defeitos nas ligações ou com efeitos de bordas, isto é, para sistemas sem invariância translacional. Usualmente essa restrição impõe enormes dificuldades em métodos tradicionais. Para diagonalizar cadeias com um número maior de sítios, implementamos um procedimento que seleciona os estados de mais baixa energia para serem usados na base de vetores do passo seguinte. Com esse tipo de truncamento de estados, fomos capazes de obter o estado fundamental e alguns estados de baixa energia de cadeias com mais de uma centena de sítios, com precisão de até cinco algarismos significativos. Nossos resultados reproduzem os da literatura para os casos conhecidos, em geral sistemas homogêneos. As aproximações desenvolvidas recentemente no contexto da Teoria do Funcional da Densidade, aplicada ao modelo de Heisenberg, e que também se aplicam a sistemas inomogêneos, estão em conformidade com nossos resultados numericamente exatos. Generalizamos o método para diagonalizar escadas de spins 1/2. Calculamos o estado fundamental e o gap de energia desse sistema, onde variamos a razão entre os acoplamentos ao longo das pernas da escada e ao longo dos degraus da mesma; nossos resultados são comparados com os da literatura. Apresentamos também a implementação do método iterativo no modelo de Hubbard, que descreve um sistema de spins itinerantes. Sabe-se que no regime de alta repulsão Coulombiana entre os spins e densidade um (número de spins igual ao número de sítios da cadeia), esse modelo é mapeado no modelo de Heisenberg, resultado que é verificado numericamente em nosso procedimento por meio do cálculo de energias de ambos os modelos em um regime paramétrico apropriado. / In this Thesis we develop a numerical method to diagonalize the Heisenberg model iteratively. In essence, we diagonalize spin chains in steps, each one corresponding to an addition of a spin to a smaller chain. The basis vectors to calculate the Hamiltonian of a N-spin chain, HN, is built by means of the direct product of the eigenvectors of the (N-1)-spin Hamiltonian, diagonalized on the previous step, by the eigenstates of the N-th added spin. Besides the common use of the conservation of the z-component of the total spin, Sz, we also exploit the conservation of the squared total spin, S2. For a specific class of spin systems we also implemented the reflection symmetry. We obtain the entire energy spectrum of spin-1/2 chains up to 20 sites, for which we show the temperature dependence of the magnetic susceptibility and specific heat, for systems with substitutional impurity spins, bond defects, border effects, i.e., for systems without translational invariance. This normally imposes enormous restrictions in many traditional methods. In order to diagonalize chains with a larger number of sites we implemented a procedure that selects lower energy states to be used in the basis vector on the next step. Using this truncation scheme, we are able to obtain low-lying energy states for chains with more than a hundred sites, up to five significant figures of accuracy. Our results reproduce those of the literature for the known cases, in general homogeneous systems. The approaches recently developed in the context of Density Functional Theory to the Heisenberg model, which also apply to inhomogeneous systems, are consistent with our numerical results. We generalize the method to diagonalize spin-1/2 ladders. We calculate the ground-state and the energy gap of this system, for arbitrary ratio of the couplings along the lags or over the rungs of the ladder. We also present the implementation of our iterative method to the Hubbard model, which describes a system of itinerant spins. It is known that in the regime of high Coulomb repulsion between the spins and unitary density (number of spins equal to the number of sites in the chain), this model is mapped onto Heisenberg one, a result which is verified numerically in our procedure by calculating the energy spectrum of both models in na appropriated parametric regime.

This thesis deals with the application of stochastic claims reserving methods to given data with some trends in claims development. It describes the chain ladder method and the generalized linear models as its stochastic framework. Some simple functions are suggested for smoothing the origin and development period coefficients from the estimated model. The extrapolation is also considered for estimation of the unobserved tail values. The residual bootstrap is used for the reparameterized model in order to get the predictive distribution of the estimated reserve together with its standard deviation as a measure of volatility. Solvency capital requirement in one year time horizon is also calculated. 1

A presente pesquisa trata da compreensão e caracterização do papel do design dentro de uma empresa de desenvolvimento de produtos digitais por meio do ponto de vista de membros das equipes de projetos e seus gestores. Uma das principais motivações para a condução do estudo provém do fato de que poucas pesquisas do tipo foram realizadas em âmbito regional e direcionadas a projetos digitais, a fim de explorar o tema. Pretende-se com a pesquisa, fornecer subsídios para um melhor aproveitamento dos benefícios do uso do design e das habilidades dos designers. A investigação foi conduzida, a partir de um estudo de caso realizado na empresa onde a pesquisadora atua também como designer e fundamentada sob uma perspectiva temporal dos papéis do design e do entendimento sobre o funcionamento de equipes multidisciplinares no processo de desenvolvimento de produtos digitais. Foram selecionados dois instrumentos de fundamentação teórica para análise dos resultados, a Design Ladder e a Taxonomia dos Papéis do Design. Por fim, apresenta-se uma discussão sobre a percepção do design pelos membros da equipe e concluiu-se que o nível de maturidade em design detectado no objeto de estudo encontra-se no segundo nível da Design Ladder: design como provedor de diferencial estético, embora seja percebido através dos discursos o desejo pelo terceiro nível, que trata o design como processo. / This research deals with the understanding and characterization of the role of design in a digital product development company and through the point of view of members of the project teams and their managers. One of the main motivation for conducting this study is due to the fact that little research has been conducted at the regional level and oriented for digital products, in order to explore the topic, and also with the aim of providing subsidies to better exploit the benefits of the use of design and designers skills. The investigation was conducted from a case study in the company where the researcher works as a designer too and based on a temporal perspective of the roles of design and understanding of the functioning of multidisciplinary teams in the digital product development process. Two instruments were selected from the theoretical background in order to analyze the results: the Design Ladder and the Taxonomy of Roles Design. Lastly, a discussion is presented on the perception of design and was concluded that the ripeness level of design are in the second level of Design Ladder: design as esthetics, although it is perceived by the team speech a desire for reach the third level who treats design as a process.

Le champ électromagnétique térahertz (THz) se situe dans l'intervalle de fréquence entre l'infrarouge et les micro-ondes, à peu près entre 1 THz à 10 THz. Ce domaine est hautement souhaitable tant pour la recherche fondamentale que pour les applications. Pourtant des sources THz compacts et accordables ne sont pas encore disponibles. Depuis la première proposition en 1970, les superréseaux semiconducteurs, dans lequel deux couches semi-conductrices atomiques avec bande interdite différente sont disposés périodiquement, fournissent de nouvelles possibilités. De nouvelles techniques et de nouveaux dispositifs deviennent réalisables. Dans cette thèse, les oscillations de Bloch dans des mini-bandes électroniques d’un superréseau polarise et la dispersion du gain associée sont utilisées pour réaliser une source THz compacte et accordable : l’oscillateur de Bloch THz. Un premier ensemble de dispositifs utilisent des réseaux dopes spécifiquement conçus pour éviter la formation de domaine d’accumulation de charges. Ces dispositifs utilisent une surface semi-isolante ou deux surfaces métalliques permettant un guidage par plasmon de surface. Cependant, malgré la réalisation de couplage par les bords ou par un réseau diffractant en surface et des mesures directes ou avec un interféromètre a transformation de Fourrier (FTIR), l’électroluminescence a été observée dans le domaine térahertz, avec un gain qui n’a pas pu etre relie aux oscillations de Bloch. Avec des superréseaux non dope, l'émission THz des oscillations de Bloch a été détectée par spectroscopie dans le domaine temporel. La dépendance de la fréquence d’émission avec le champ électrique appliqué constitue une preuve directe des oscillations de Bloch. L’échelle de Wannier Stark des trous sous pompage optique continu a aussi été observe dans les superréseaux non dopes. Avec l’augmentation de la puissance de pompage optique, les pics du photocourant se décalent et leurs formes deviennent asymétriques. L’évolution est attribue a l’accumulation des porteurs photogénérés dans les deux couches encadrant le superréseau. En outre, pour une puissance de pompage élevée, la bistabilité du photocourant a été également observée. / Terahertz (THz) electromagnetic field, which lies in the frequency gap between the infrared and microwave, roughly between 1 THz to 10 THz, is highly desirable for both fundamental research and application. Yet tuneable compact THz sources are still not available. On the other hand, ever since first proposed in 1970, semiconductor superlattice provides new playground for various new technique and devices of tremendous research and application interest. In this thesis, an innovative theme, relying on Bloch oscillations in a dc biased semiconductor superlattice, is explored to realize tunable compact THz source THz Bloch oscillator. For doped superlattice Bloch oscillator, we designed quantum cascade super-superlattice structure to realize Bloch oscillations whilst prohibit electrical domain formation. The designed structures were processed into various waveguide and grating devices for electroluminescence detection using Fourier transform infrared spectroscopy (FTIR). The Bloch gain of semi-insulating surface plasmon waveguide device was also measured using THz time domain spectroscopy. Even though the electroluminescence and gain at THz regime were observed, no direct evidence of Bloch emission was confirmed. For undoped superlattice, the THz emission from Bloch oscillations was observed by time domain spectroscopy. At last, the photocurrent corresponding to heavy hole and Wannier Stark Ladder (WSL) states transitions in undoped superlattice was studied. Under CW laser pumping, the photocurrent as function of the applied voltage showed multiple WSL peaks, which indicated laser induced and controllable negative differential conductance (NDC). With increasing pumping power, the nonlinear NDC regime and bistable states were investigated as well.

The perceptions of elementary teachers with regard to the leadership behavior exhibited by their principals and to the organizational climate of their schools were examined in this study. The purpose of the study was to determine whether Tennessee elementary principals who achieved Career Ladder III standing exhibited more effective leadership behaviors and maintained a more suitable organizational climates than Career Ladder I principals. This study followed the ex-post facto research approach and utilized data obtained through use of the Leadership Behavior Description Questionnaire Form 12 (LBDQ-12) and the Revised Organizational Climate Description Questionnaire (OCDQ-RE). Responses were obtained from 590 teachers who represented 26 randomly selected elementary schools in northeast Tennessee, 11 of which were administered by Career Ladder III principals and 16 administered by Career Ladder I principals. Analysis of Variance (ANOVA) was applied to the data to determine significance at the.05 level. ANOVA was selected because it permitted the researcher to evaluate the mean differences in perceived leadership behavior and organizational climate simultaneously while maintaining the Type I error rate at the preestablished.05 significance level for the entire set of comparisons. No significant differences were found in total leader behavior or in any dimension of leader behavior, as measured by the LBDQ-12, for Career Ladder III elementary principals when compared to Career Ladder I elementary principals. No significant differences were found in any dimension of organizational climate, as measured by the OCDQ-RE, for elementary schools administered by Career Ladder III principals when compared to elementary schools administered by Career Ladder I principals. Recommendations for future research were given.

The problem of this study was to identify and discriminate the perceptions of administrators, Career Ladder teachers, and non-ladder teachers in the Tennessee public schools toward the Tennessee Career Ladder Teacher Evaluation System. Demographic data and information were collected by a state-wide survey of Tennessee educators. Two research questions were raised to guide the study. The comparison data collected pertained to sex and age of the participants, highest degree completed by the participants, participants' Career Ladder status, professional membership, professional experience, and the type and classification of school in which the participants worked. The information concerning educators' perceptions regarding the Career Ladder Program was obtained through their responses to the 30 statements in the research instrument which dealt with the various important aspects of the program. Major findings indicated that significant differences in perception regarding the Career Ladder Program existed among the Tennessee educators. In general, teachers who have obtained Levels II and III status on the Career Ladder and administrators perceived the program positively while Level I teachers and particularly non-ladder teachers tended to perceive it rather negatively. No groups surveyed felt that the evaluation process was well-understood or that the program encouraged diversity in teaching behavior. No groups felt that differences in learners, schools, and school systems were considered when assessing the effectiveness of teaching behavior under the current evaluation system. All educators felt that need for reducing the amount of paperwork required in teacher preparation for evaluation, and were aware of the "gamesmanship" dimension of the program. There was agreement among all groups in the study that the Career Ladder Program had failed to attract the best people into the teaching profession, failed to retain them, and has done little to enhance the teacher's public esteem. Despite the agreements, it is evident that Tennessee public school teachers and administrators held different perceptions toward the Career Ladder Program. A relationship seems to exist between administrators and higher level teachers and a more positive perception concerning the program. On the other hand, it appears to be true that lower lever teachers and non-ladder teachers are associated with a generally negative perception regarding the Career Ladder Program. (Abstract shortened with permission of author.)