Studiens syfte är att belysa begreppet barn i behov av särskilt stöd. Detta har vi granskat genom att undersöka förskolepersonalens uppfattningar om vilka barn som kan tänkas vara i behov av särskilt stöd och hur arbetet med dessa barn i förskolan ser ut, samt vilket stöd dessa barn kan få. Vi har använt oss av en kvalitativ metod och använt oss utav semi-strukturerade intervjuer och de respondenterna som deltog representerades av verksamma pedagoger i förskolan. Vi har sedan analyserat intervjustudien och studiens resultat tyder på att begreppet barn i behov av särskilt stöd är svårdefinierat och otydlig. Resultatet visade även att en diagnos i många fall kan underlätta arbetet för pedagogerna och de barn som är i behov av särskilt stöd, eftersom de lättare kan få tillgång till det stöd som behövs, exempelvis en resurs. Vår studie visade även på att pedagogernas bemötande, kunskap och kompetens om de barn som är i behov av särskilt stöd är viktigt för att kunna stötta barnen. Slutsatsen av vår intervjustudie blir att barn i behov av särskilt stöd är ett komplext område att förhålla sig till samt att, det krävs fördjupade kunskaper inom de olika områden som finns gällande barn i behov av särskilt stöd. Exempelvis fördjupade kunskaper om diagnoser, funktionsnedsättningar, språksvårigheter samt verktygens användningsområden. Slutsatsen blir även att tillgången till resurser är begränsad hos de förskolor som vi har besökt och det krävs oftast en diagnos eller uttalad funktionsnedsättning, för att pedagogerna i förskolans verksamhet ska få in resurser till ett barn som är i behov av särskilt stöd.

L'objet de cette thèse est l'étude d'algorithmes itératifs permettant de résoudre des problèmes d'optimisation convexe avec ou sans contraintes fonctionnelles, des problèmes de résolutions d'inégalités variationnelles à opérateur monotone et des problèmes de recherche de point selle. Ces problèmes sont envisagés lorsque la dimension de l'espace de recherche est grande et lorsque les valeurs des différentes fonctions étudiées et leur sous/sur-gradients ne sont pas connues exactement et ne sont accessibles qu'au travers d'un oracle stochastique. Les algorithmes que nous étudions sont des adaptations au cas stochastique de deux algorithmes : le premier inspiré de la méthode de descente en miroir de Nemirovski et Yudin et le second, de l'algorithme d'extrapolation duale de Nesterov. Pour chacun de ces deux algorithmes, nous donnons des bornes pour l'espérance et pour les déviations modérées de l'erreur d'approximation sous différentes hypothèses de régularité pour tous les problèmes sans contraintes fonctionnelles envisagées et nous donnons des versions adaptatives de ces algorithmes qui permettent de s'affranchir de connaître certains paramètres de ces problèmes non accessibles en pratique. Enfin nous montrons comment, à l'aide d'un algorithme auxiliaire inspiré de la méthode de Newton et des résultats obtenus lors de la résolution des problèmes de recherche de point selle, il est possible de résoudre des problèmes d'optimisation sous contraintes fonctionnelles.

Bakgrund: Redovisnings- och företagsskandaler som ägde rum på 2000-talet i USA och flera europeiska länder satte fart på bolagsstyrningsdebatten. Många länder började införa regler med syfte att förbättra bolagsstyrningen och öka förtroendet för näringslivet. Sverige drabbades liksom andra länder av redovisningsskandaler, och som följd till detta infördes den svenska koden för bolagsstyrning för att återuppbygga marknadens förtroende för de svenska bolagen. Första juli 2005 tillämpades koden i Stockholmsbörsen för alla bolag på A-listan och O-listan med börsvärde över 3 miljarder kronor det vill säga de bolag som ingick i kategorin Large Cap.Syfte: Syftet med denna uppsats är att undersöka och analysera hur bolag på OMX Stockholmsbörsen, Large Cap, avviker från koden och om de avviker från samma punkter i koden samt hur utförligt förklarar dessa bolag avvikelserna. Med hjälp av vetenskapliga artiklar som teoretiskt perspektiv vill jag få förståelse för varför koden finns och varför bolag avviker från den samt undersöka om koden fungerar som det var tänkt.Metod: För att undersöka i vilken omfattning bolag avviker från koden och vilka punkter det som bolagen avviker från har jag läst igenom och granskat bolagens årsredovisningar och bolagsstyrningsrapporter.Teoretisk referensram: I detta kapitel har jag tagit upp tidigare forskning som handlar om andra länders bolagskoder och principen följa eller förklara, samt orsaker till varför bolag väljer att följa eller avvika från koden. Dessa studier är gjorda i Tyskland och Storbritannien.5Empiri: Det empiriska materialet består av årsredovisningar och bolagsstyrningsrapporter för de undersökta 28 bolagen inom kategorin Large Cap. Jag har undersökt de senaste åren när jag gjorde undersökningen, dvs. 2010 och 2011.Slutsatser: Det jag kom fram till i min studie är att ägarstrukturen är en bidragande faktor till varför bolag väljer att avvika från koden. Bolag med ett koncentrerat ägande tenderar att avvika från koden. Dessutom har jag märkt att dessa bolag ägs av Industrivärden, Investor. De bolagen kännetecknas även av att de har starka familjeägare som äger mer än 30 % av aktierna, dessa ägare vill oftast ha kontroll över bolaget och därför avviker de från den svenska koden. / Background: Accounting and corporate scandals that took place in the 2000s in the United States and several European countries took off in the corporate governance debate. Many countries began to adopt rules designed to develop corporate governance and increase confidence in the business world. Sweden suffered like other countries of accounting scandals, and as a result of this came the Swedish Code of Corporate Governance to rebuild market confidence to the Swedish companies. The first of July 2005 the code was applied in the Stockholm Stock Exchange for all companies on the A-list and the O-list with market capitalization exceeding 3 billion i.e. companies included in the Large Cap. Purpose: The purpose of this study is to examine and analyze how companies on the OMX Stockholm Stock Exchange, Large Cap, deviates from the Swedish code and if they deviate from the same points in the code and how thoroughly these companies explains their deviations. I want to get an understanding of why the code exists and why the companies deviate from it and whether the code works as expected. Methodology: To examine the extent to which companies deviate from the code and which points that the companies differ from, I have studied and observed the companies' annual reports and corporate governance reports. Theoretical Framework: In this chapter, I have raised earlier research that deals with other codes and the principle comply or explain, and the reasons why companies choose to follow or deviate from the code. These studies took place in Germany and the UK.7Empirical foundation: The empirical data consists of annual reports and corporate governance reports of the surveyed 28 companies in the Large Cap from both 2010 and 2011. Conclusions: The conclusion of this study is that the ownership structure is a contributing factor to why companies choose to deviate from the code. Companies with a concentrated ownership tend to deviate from the code. I have also noticed that these companies are owned by Industrivärden and Investor. The companies are also characterized by having strong family owners who own more than 30% of the shares, the owners usually want to have control over the company and that are why they differ from the Swedish Code.

Magistro darbe sprendėme diferencialinę difuzijos lygtį naujais metodais. Išanalizavę standartinius kintamo difuzijos koeficiento parabolinių lygčių sprendimo metodus, mes šiame darbe pasiūlėme spręsti šias lygtis naudojant vadinamąsias „tilto“ funkcijas. Išbandėme dviejų rūšių „tilto“ funkcijas: hiperbolinio tangento ir trigonometrinio. Diferencialinės lygties sprendinio ieškojome per „tilto“ funkcijų ir polinomų sandaugų sumą: trigonometrinei „tilto“ funkcijai ir hiperbolinei tangento „tilto“ funkcijai. Gavome kompiuterinius sprendinius ir nustatėme tų sprendinių paklaidas. Palyginę trigonometrinio bei hiperbolinio tangento „tilto“ funkcijos paklaidų standartinius nuokrypius gavome, kad tikslesnis yra hiperbolinio tangento „tilto“ funkcijos metodas. / Master thesis solved differential equation of diffusion of new techniques methods. Having analyzed the standard variable diffusion coefficient parabolic equation solution methods suggested in this work we solve these equations using the so-called "bridge" function. Tried two types of "bridge" functions: tangent hyperbolic and trigonometric. Differential equation, the solution we were looking for a "bridge" function and the amount of products of powers of polynomials: trigonometry "bridge" function and hyperbolic tangent of a "bridge" function. We have received computer-based solutions and the solutions found at the margins. A comparison of hyperbolic tangent trigonometric "bridge" function of the error standard deviations have received, the more accurate the hyperbolic tangent of a "bridge" function approach.

Les risques sont inhérents à l’activité industrielle. Les prévoir et les maîtriser sont essentiels pour la conception et la conduite en sécurité des procédés. La réglementation des risques majeurs impose aux exploitants la réalisation d’études de sécurité quantitatives. La stratégie de maîtrise des risques repose sur la pertinence des analyses de risques. En marche dégradée, la dynamique des événements est déterminante pour quantifier les risques. Toutefois, de nos jours cette connaissance est difficilement accessible. Ce travail propose une méthodologie d’analyse de risques quantitative qui combine la méthode HAZOP, le retour d’expérience et la simulation dynamique de dérives de procédés. Elle repose sur quatre grandes étapes : La première étape est l’étude du fonctionnement normal du procédé. Pour cela, le procédé est décrit de façon détaillée. Des études complémentaires de caractérisation des produits et du milieu réactionnel sont menées si nécessaires. Ensuite, le procédé est simulé dynamiquement en fonctionnement normal. Lors de la seconde étape, parmi les dérives définies par l’HAZOP et le retour d’expérience, l’analyste discrimine celles dont les conséquences ne sont pas prévisibles et/ou nécessitent d’être quantifiées. La troisième phase fournit une quantification du risque sur la base de la simulation dynamique des scenarii retenus. Lors de la dernière étape, des mesures de maîtrise des risques sont définies et ajoutées au procédé lorsque le niveau de risque est supérieur au risque tolérable. Le risque résiduel est ensuite calculé jusqu’à l’atteinte de la cible sécurité. Le logiciel Aspen Plus Dynamics est sélectionné. Trois études de cas sont choisies pour démontrer d’une part, la faisabilité de la méthodologie et d’autre part, la diversité de son champ d’application : · la première étude de cas porte sur un réacteur semi-continu siège d’une réaction exothermique. L’oxydation du thiosulfate de sodium par le peroxyde d’hydrogène est choisie. Ce cas relativement simple permet d’illustrer la diversité des causes pouvant être simulées (erreur procédurale, défaut matériel, contamination de produits, …) et la possibilité d’étudier des dérives simultanées (perte de refroidissement du milieu et sous dimensionnement de la soupape de sécurité). · le deuxième cas concerne un réacteur semi-batch dans lequel une réaction exothermique de sulfonation est opérée. Elle est particulièrement difficile à mettre en œuvre car le risque d’emballement thermique est élevé. Cette étude montre l’intérêt de notre approche dans la définition des conditions opératoires pour la conduite en sécurité. · le troisième cas d’étude porte sur un procédé continu de fabrication du propylène glycol composé d’un réacteur et de deux colonnes de distillation en série. L’objectif est ici d’étudier la propagation de dérives le long du procédé. Sur la base du retour d’expérience, deux dérives au niveau du rebouilleur de la première colonne sont étudiées et illustrent les risques de pleurage et d’engorgement. La simulation dynamique illustre la propagation d’une dérive et ses conséquences sur la colonne suivante. / Risks are inherent to industrial activity. Predicting and controlling them is essential to the processes design and safe operation. Quantitative safety studies are imposed by the major hazard regulations. The risk management strategy relies on the relevance of risk analyzes. In degraded conditions, the dynamics of events are decisive for risks quantification. However, nowadays this knowledge is a real challenge. This work proposes a methodology of quantitative risk analysis, which combines the HAZOP method, the lessons learned from previous accidents and the dynamic simulation of process deviations. It is based on four main stages: The first stage is the study of the process normal operation. For this, the process is described in detail. Additional studies to characterize the products and the reaction are carried out if necessary. Then, the process is dynamically simulated in normal operation conditions. During the second step, among all the deviations defined by the HAZOP and lessons learned, the analyst discriminates those whose consequences are not predictable and/or need to be quantified. The third phase provides a risk quantification based on the dynamic simulation of the selected scenarios. In the last step, safety barriers are defined and added to the process when the risk level is greater than the tolerable risk. The residual risk is then calculated until the safety target is reached. Aspen Plus Dynamics software is selected. Three case studies are chosen in order to demonstrate, on the one hand, the feasibility of the methodology and, on the other hand, the diversity of its scope: · the first case study is a semi-continuous reactor with an exothermic reaction study. The oxidation of sodium thiosulfate by hydrogen peroxide is selected. This relatively simple case illustrates the diversity of causes that can be simulated (procedural error, material defect, product contamination …) and the possibility of studying simultaneous deviations (loss of cooling and under sized safety valve for example). · the second case concerns a semi-batch reactor in which an exothermic reaction of sulphonation is carried out. This reaction is particularly difficult to conduct because of the thermal runaway high risk. This study shows our approach’s interest in the definition of the operating conditions for safe operation. · the third case study concerns a continuous process of propylene glycol production. It is composed of a reactor and two distillation columns in series. The objective is to study the propagation of deviations along the process. Based on lessons learned, two deviations in the first column reboiler are studied and illustrate the flooding and weeping risks. Dynamic simulation illustrates the propagation of a deviation and its consequences on the second column

O propósito deste trabalho foi refletir sobre o atendimento em grupo com crianças autistas. O autismo é aqui compreendido não pelo viés do que lhe falta, mas como um sujeito em potencial que apresenta dificuldades em três eixos principais, a saber: o eixo da esfera relacional, o eixo da linguagem e o eixo do repertório de interesses e das estereotipias. À luz da psicanálise de orientação lacaniana, o que interessa no autismo não é o diagnóstico em si, mas o jogo pulsional que se desenha interrompido e a possibilidade de intervenção que preconiza o sujeito como possível de se constituir. Entre as estratégias plausíveis de intervenção, o grupo tem lugar de realce nos manejos do trabalho clínico com crianças autistas, apesar de existir, no autismo, uma séria dificuldade na relação com o outro. Portanto, esta pesquisa se propôs a analisar o dispositivo terapêutico grupal como um recurso que contribui no processo de constituição subjetiva dessas crianças. A pesquisa foi realizada mediante coleta do depoimento de quatro profissionais de diferentes instituições da cidade do Recife, as quais foram selecionadas por serem especializadas no tratamento de crianças autistas e possuírem larga experiência e reconhecimento tanto social quanto acadêmico. Para análise do conteúdo dos depoimentos, foi feita uma categorização das respostas de modo a obter uma compreensão de cada depoimento per si e, ao mesmo tempo, uma comparação entre eles, verificando os pontos convergentes e divergentes. De forma geral, os pontos divergentes diziam respeito mais ao modo específico de funcionamento de cada grupo do que ao objetivo do grupo em si. Os aspectos mais realçados no trabalho de grupo foram: formação do grupo, planejamento, investimento libidinal do terapeuta, linguagem/comunicação, transferência, reunião de equipe, janela pulsional e saída do grupo. Tais aspectos não só viabilizam esse recurso clínico como auxiliam na [re]estruturação psíquica de crianças autistas. Portanto, é consenso que o dispositivo terapêutico grupal é uma estratégia de intervenção possível e pode provocar rearranjos psíquicos que são postos a serviço da constituição subjetiva e, sendo assim, do sujeito.

Made available in DSpace on 2014-12-17T15:26:39Z (GMT). No. of bitstreams: 1 MariaJS_DISSERT.pdf: 2706687 bytes, checksum: 2f98eddad7bbc278c03ee45e4e226d95 (MD5) Previous issue date: 2013-04-24 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / This paper we study a random strategy called MOSES, which was introduced in 1996 by Fran?cois. Asymptotic results of this strategy; behavior of the stationary distributions of the chain associated to strategy, were derived by Fran?cois, in 1998, of the theory of Freidlin and Wentzell [8]. Detailings of these results are in this work. Moreover, we noted that an alternative approach the convergence of this strategy is possible without making use of theory of Freidlin and Wentzell, yielding the visit almost certain of the strategy to uniform populations which contain the minimum. Some simulations in Matlab are presented in this work / Neste trabalho estudamos uma estrat?gia aleat?ria chamada de MOSES, que foi introduzida por Fran?ois em 1996. Resultados assint?ticos desta estrat?gia; comportamento das distribui??es estacion?rias da cadeia associada a estrat?gia, foram derivados por Fran?ois, em 1998, da teoria de Freidlin e Wentzell [8]. Detalhamentos destes resultados est?o neste trabalho. Por outro lado, notamos que uma abordagem alternativa da converg?ncia desta estrat?gia ? poss?vel sem fazer uso da teoria de Freidlin e Wentzell, obtendo-se a visita quase certa da estrat?gia as popula??es uniformes que cont?m o m?m?nimo. Algumas simula??es no Matlab s?o apresentadas neste trabalho

Desvios da norma padrão, comumente chamados de \"erros\", são fatos comuns na escrita dos aprendizes da variante culta de uma língua materna como o português brasileiro. Tratados como um \"mal a ser combatido\", eles são, na verdade, importantes indícios do processo de assimilação da escrita culta pelo falante nativo. Revelam qual a tendência da transformação que naturalmente ocorre numa língua, demonstrando, por exemplo, a obsolência das gramáticas tradicionais, que não aceitam determinadas construções já muito freqüentes. Mas seria possível detectar algum padrão desses desvios? Haveria desvios típicos de um determinado perfil de redatores? Essas indagações motivaram a presente investigação, que se baseou na concepção de que esses estudos são de natureza empírica, comprometidos com a noção de que a língua funciona tal como um sistema probabilístico, de onde é possível prever tendências, por exemplo, de mudança. Falamos, pois, de uma investigação à luz da Lingüística de Corpus. Composto de redações do Exame Nacional do Ensino Médio (ENEM), edição de 2002, cedidas pelo Instituto Nacional de Estudos e Pesquisas Educacionais (INEP) juntamente com determinados traços do perfil dos redatores, construímos um corpus que foi batizado de Corvo, e se ocupou de uma faixa específica de textos: a de pior desempenho no ENEM no quesito domínio da norma culta. Observamos, desse modo, textos em que, supostamente, há freqüência maior de desvios e maior variedade de tipos de desvios. Nossa metodologia de pesquisa apoiou-se no uso do revisor gramatical automático ReGra, bastante popular no país e que auxilia o usuário no uso correto do português culto padrão. Além disso, construímos um material próprio de detecção e classificação dos desvios gramaticais, aumentando a capacidade de tratamento automático dos textos. Assim, foi possível gerar uma versão do corpus anotada em desvios, i.e., os textos apresentam indicações de quando e qual tipo de desvios ocorrem. Como resultado temos um mapeamento do Corvo; ou seja, um panorama dos desvios típicos de um determinado tip o de perfil de redator. Constatamos a deficiência ortográfica como o traço típico do grupo de indivíduos investigado, mas, sobretudo, que a ortografia é motor para o pleno funcionamento de uma revisão gramatical automática. O revisor ReGra mostrou-se incapaz de processar satisfatoriamente textos desse tipo de redator, mas, ainda assim, comprovou que esses textos apresentam desvios gramaticais de tratamento complexo, cuja intervenção do revisor, se acontece, pouco altera na qualidade geral dos mesmos. Com respeito à tipologia de desvio, pudemos constatar a validade da tipologia aplicada na pesquisa, que advém do ReGra e, portanto, está à margem das discussões teóricas ortodoxas. De fato, há recorrência de tipos de desvios, e isso numa freqüência que nos autoriza admitir a fraca assimilação de certas regras gramaticais tomadas como básicas (p.ex., a pontuação, a concordância e a regência). Constatamos, com relação ao perfil de redatores, que textos com maior potencial para a revisão da escrita, i.e., aqueles que alteram significativamente a qualidade textual com interferências pontuais de revisão, são justamente os produzidos pelos concluintes do ensino médio e não pelos egressos. / Deviations from the standard norm, usually called \'mistakes\', are common events in writing pieces of language learners speakers of a native language such as Brazilian Portuguese. They are treated as \'an evil that must be fought\'. They are, in fact, important evidence of the acquisition process of writing in the standard norm by the native speaker. They reveal the transformation trend, which normally occurs in a language, showing, for instance, the obsolescence of traditional grammars that do not accept certain patterns, which are frequent nonetheless. However, is it possible to identify a pattern in these deviations? Are there common deviations among a certain profile of students? These are the questions motivating this study, which is based on the concept that these investigations are empirical in nature, and are marked by the notion that language operates as a probabilistic system, in which it is possible to forecast trends of change, for example. We are, therefore, speaking of an investigation in the light of Corpus Linguistics. We compiled a corpus of essays written during the National Middle Education Exam (ENEM) carried out in 2002. These essays were obtained with the National Institute for Research in Education (INEP) together with the profile of the students. The corpus was called Corvo, and it is made up of a certain bracket of texts: those having obtained the worst performance rate in the ENEM in the standard norm category. We observed, therefore, texts in which there are, allegedly, a greater frequency of deviations and a greater variety of kinds of deviations. The research methodology was supported by the electronic grammar checker - ReGra - which is very popular in the country, and helps the user in writing standard Portuguese correctly. In addition, we built a specific tool for identifying and classifying grammar deviations, thus, increasing the ability to treat the texts electronically. Therefore, it was possible to generate an annotated version of the corpus according to the deviations, i.e., the texts were annotated according to when and what kind of deviations they presented. As a result, we have a mapping of the Corvo; that is, a view of the common deviations of students belonging to a certain profile. We identified poor spelling as a common feature of the group, but, above all, that spelling is the engine enabling a full grammatical check to operate. The ReGra grammar checker was not able to satisfactorily process these kinds of texts, but, even so, it proved that these texts presented complex grammar deviations, and the intervention of the checker, when it is applied, little alters their overall quality. In regard to deviation typology, we identified the validity of the typology used in this study, which results from the ReGra and, therefore, lies in the outskirts of orthodox theoretical discussions. In fact, certain kinds of deviations reoccur at a frequency that enables us to admit a poor assimilation of certain grammatical rules considered basic (e.g., punctuation, agreement and use of prepositions). We found that in regard to the profile of the students, texts with a greater writing check potential, that is, a check that would significantly improve text quality through individual checking interferences, are te xts produced by students who finished middle education and not those produced by students who are finishing the course.

Seja $\\Sigma_(\\mathbb)$ um shift enumerável topologicamente mixing com a propriedade BIP sobre o alfabeto $\\mathbb$, $f: \\Sigma_(\\mathbb) ightarrow \\mathbb$ um potencial com variação somável e pressão topológica finita. Sob hipóteses adequadas provamos a existência de um princípio dos grandes desvios para a familia de estados de Gibbs $(\\mu_{\\beta})_{\\beta > 0}$, onde cada $\\mu_{\\beta}$ é a medida de Gibbs associada ao potencial $\\beta f$. Para fazer isso generalizamos alguns teoremas de Otimização Ergódica para shifts de Markov enumeráveis. Esse resultado generaliza o mesmo princípio no caso de um subshift topologicamente mixing sobre um alfabeto finito, previamente provado por A. Baraviera, A. Lopes e P. Thieullen. / Let $\\Sigma_(\\mathbb)$ be a topologically mixing countable Markov shift with the BIP property over the alphabet $\\mathbb$ and a potential $f: \\Sigma_(\\mathbb) ightarrow \\mathbb$ with summable variation and finite pressure. Under suitable hypotheses, we prove the existence of a large deviation principle for the family of Gibbs states $(\\mu_{\\beta})_{\\beta > 0}$ where each $\\mu_{\\beta}$ is the Gibbs measure associated to the potential $\\beta f$. For do this we generalize some theorems from finite to countable Markov shifts in Ergodic Optimization. This result generalizes the same principle in the case of topologically mixing subshifts over a finite alphabet previously proved by A. Baraviera, A. Lopes and P. Thieullen.

Cette thèse est consacrée à l’étude de théorèmes limites : grandes déviations et déviations modérées pour des estimateurs liés à des modèles financiers. Dans une première partie, nous nous sommes intéressés à l’étude des déviations grandes et modérées des estimateurs de la covariation et de la (co)volatilité réalisée issus des fonctionnelles associées à deux processus de diffusion couplés de manière synchronisée. Les techniques utilisées dans ces travaux sont basées d’une part sur celles utilisées dans Djellout-Guillin-Wu et sur la sous additivité et sur la notion d’approximation exponentielle inspirées des travaux de J. Najim d’autre part. Dans une deuxième partie, on considère que les deux processus de diffusion sont observés de manière non synchronisée et on établit des déviations modérées pour l’estimateur de la variation généralisée et pour celui de Hayashi-Yoshida. Les résultats sont obtenus par l’utilisation d’une nouvelle approche sur les déviations modérées des variables aléatoires m−dépendantes vérifiant des conditions de type "Chen-Ledoux". Dans la troisième et dernière partie, on s’intéresse à l’étude processus autorégressif d’ordre p dont le bruit est un processus autorégressif d’ordre q. On montre des déviations modérées pour certains estimateurs associés à notre modèle dont la statistique de Durbin-Watson. Les résultats sont donnés dans le cas où le bruit est gaussien puis dans le cas de condition de type "Chen-Ledoux" portant sur le bruit. / This thesis is devoted to the study of the limits theorem : large and moderate déviations for some financial mathematicals estimators. In the first part, we studied the large and moderate deviations of the estimators of covariation and the realized (co)volatility obtained from the functional associated to two diffusion processes coupled in synchronous manner. The techniques used in this work are based, on the one hand, on those used in Djellout-Guillin-Wu and the subadditivity and the exponential approximation notion inspired by J. Najim results on the other hand. In the second part, we consider that ours two diffusion processes are observed in a nonsynchronized manner and on the establish the moderate deviations for the generalised bipower variation estimator and the Hayashi-Yoshida estimator. The results are obtained by using a new approach on the moderate deviations of the m−dependent random variables based on the Chen-Ledoux type condition. In the third and last part, we study the stable autoregressive process of order p where the driven noise is also given by a q-order autoregressive process. We prove the moderate deviations for some estimators associated with our model such as the Durbin-Watson statistic. The results are given in the case where the driven noise is the normally distributed then in the case where the driven noise satisfy a Chen-Ledoux type condition.