O crescimento acelerado da produção científica brasileira nos anos recentes foi acompanhado pela expansão das colaborações científicas domésticas. Neste estudo, olhamos mais atentamente esse assunto na tentativa pioneira de identificar padrões espaciais da produção e colaboração científica no Brasil, e avaliar o papel da proximidade geográfica na determinação das interações entre os pesquisadores brasileiros. Por meio de uma base única composta por mais de um milhão de pesquisadores registrados na Plataforma Lattes e de sete milhões de publicações científicas, coletamos e consolidamos informações sobre as colaborações científicas inter-regionais em termos de redes de coautorias entre 1.347 municípios brasileiros ao longo do período compreendido entre 1990 e 2010, o que permitiu uma abrangência de dados e perspectiva de análise inéditas na literatura. Os efeitos da distância geográfica nas redes de colaboração são mensurados para as diferentes áreas do conhecimento por meio da estimação de modelos de interações espaciais. Os principais resultados sugerem fortes evidências de um processo de desconcentração espacial da produção científica nos últimos anos associado à expansão das redes de colaboração e ao aumento da participação de autores das regiões cientificamente menos tradicionais, tais como Sul e Nordeste. Ademais, também encontramos evidência de que a distância ainda desempenha papel crucial na determinação da intensidade dos fluxos de conhecimento nas redes de colaboração científica no Brasil, embora a magnitude do efeito varie entre as redes das diferentes áreas do conhecimento. Por exemplo, verificamos que o distanciamento de 200 quilômetros entre dois pesquisadores implica na redução média de 22% ou 45% na probabilidade de haver colaboração entre eles, caso eles sejam de Linguística, Letras e Artes ou Ciências Exatas e da Terra, respectivamente. / Recent years have witnessed an accelerated growth of Brazilian scientific production that was accompanied by an expansion of domestic research collaboration. In this paper we look more closely at this issue in a pioneering attempt to identify spatial patterns of research production and collaboration in Brazil, and to measure the role of geographical proximity in determining interaction between Brazilian researchers. Using a unique database comprised of over one million researchers registered in the Brazilian Lattes Platform and seven million scientific publications, we collect and consolidate information on interregional research collaboration in terms of co-authorship networks among 1,347 Brazilian cities over the period between 1990 and 2010, which enabled a range of data and analysis perspective unprecedented in literature. The effects of geographical distance on research collaboration are measured for different knowledge areas under the estimation of spatial interaction models. The main results suggest strong evidence of spatial de-concentration of scientific production in the last years with expansion of research collaboration networks and an increased participation of authors in scientifically less traditional regions, such as South and Northeast. Moreover, we also find evidence that distance still plays a crucial role in determining the intensity of knowledge flows in scientific collaboration networks in Brazil, although the magnitude of such effects varies among networks of different knowledge areas. For instance, we found that the distancing of 200 kilometers between two researchers implies an average reduction of 22% or 45% on probability of collaboration among them, if they are of Linguistics, Letters and Arts or Exact and Earth Sciences, respectively.

Les marchés financiers occupent une place prépondérante dans l’économie. La future évolution des législations dans le domaine de la finance mondiale va rendre inévitable l’introduction de frictions pour éviter les mouvements spéculatifs des capitaux, toujours menaçants d’une crise. C’est pourquoi nous nous intéressons principalement, ici, aux modèles de marchés financiers avec coûts de transaction.Cette thèse se compose de trois chapitres. Le premier établit un critère d’absence d’opportunité d’arbitrage donnant l’existence de systèmes de prix consistants, i.e. martingales évoluant dans le cône dual positif exprimé en unités physiques, pour une famille de modèles de marchés financiers en temps continu avec petits coûts de transaction.Dans le deuxième chapitre, nous montrons la convergence des ensembles de sur-réplication d’une option européenne dans le cadre de la convergence topologique des ensembles. Dans des modèles multidimensionnels avec coûts de transaction décroissants a l’ordre n−1/2, nous donnons une description de l’ensemble limite pour des modèles particuliers et en déduisons des inclusions pour les modèles généraux (modèles de KABANOV). Le troisième chapitre est dédié a l’approximation du prix d’options européennes pour des modèles avec diffusion très générale (sans coûts de transaction). Nous étudions les propriétés des pay-offs pour pouvoir utiliser au mieux l’approximation du processus de prix du sous-jacent par un processus intuitif défini par récurrence grâce aux itérations de PICARD / Financial markets play a prevailing role in the economy. The future legislation development in the field of globalfinance will unavoidably lead to friction to prevent speculative capital movements, always threatening with crisis. Thatis why we are interested in the financial market models with transaction costs.This thesis consists of three chapters. The first one establishes a criterion of absence of arbitrage opportunitiesgiving the existence of consistent price systems, i.e. martingale evolving in the dual cone expressed in physical units.The criterion holds for a family of financial market models in continuous time with small transaction costs.In the second chapter, we show the convergence of super-replication sets for a European option in the contextof the topological convergence of sets. In multivariate models with transaction costs decreasing at rate n-1/2, we give adescription of the limit set for specific models. We deduce inclusions for general models (KABANOV's models).The third chapter is dedicated to the approximation of the European option price for models with very generaldiffusion (without transaction costs). We study properties of the pay-off to make best use of the approximation of theunderlying asset price, based on PICARD iterations.

In this note, we consider an inverse problem in a jump diffusion model. Using characteristic functions we prove the injectivity of the forward operator mapping the five parameters determining the model to the density function of the return distribution.

Das Ziel der Dissertation ist die Entwicklung statistischer Testverfahren zur Überprüfung parametrischer Modelle für die Dynamik zeitstetiger Prozesse und die Anwendung der entwickelten Methoden auf Finanzmarktdaten. Besonderes Augenmerk wird auf die statistische Methodik und die Untersuchung der Testeigenschaften in endlichen Stichproben gelegt, da diese in empirischen Untersuchungen von entscheidener Bedeutung sind. Alle Kapitel der Dissertation umfassen eine empirische Analyse, in der die vorgestellten Tests auf Finanzmarktdaten angewandt werden. / The aim of the thesis is to provide a wide range of statistical methods designed to test parametric assumptions about the evolution of continuous time processes in financial markets. The main focus is on the statistical methodology and the investigation of the properties of the proposed methods when applied to finite samples. The latter aspect is particularly important for empirical applications. All chapters include an empirical analysis of financial data using the developed methods.

V předložené práci se zabýváme problematikou stochastického modelování úro- kových sazeb. Jedním z nejobvyklejších postup· je modelovat dynamiku úroko- vých sazeb pomocí stochastické diferenciální rovnice difúze, jejímiž základními kameny jsou funkce driftu a funkce difúze. Od 70. let 20. století byla navržena celá řada model· tohoto typu, a ačkoli se tyto modely neustále zdokonalují, vyvstává přirozená otázka, zda se historicky pozorované úrokové sazby skutečně takovými difúzními rovnicemi řídily. V této práci budeme právě uvedenou hypo- tézu testovat pro několik nejběžnějších jednofaktorových model· úrokové sazby první generace. Z historických dat odhadneme obecnou momentovou metodou a metodou maximální věrohodnosti parametry jednotlivých difúzních rovnic a následně provedeme statistické testy dobré shody proložení těchto rovnic pozo- rovanými daty. 1

Submitted by William Justo Figueiro (williamjf) on 2015-07-09T21:57:13Z No. of bitstreams: 1 22b.pdf: 2481404 bytes, checksum: baa76e5fa7d1e919420ae0df3e76ca46 (MD5) / Made available in DSpace on 2015-07-09T21:57:13Z (GMT). No. of bitstreams: 1 22b.pdf: 2481404 bytes, checksum: baa76e5fa7d1e919420ae0df3e76ca46 (MD5) Previous issue date: 2013-07-26 / Nenhuma / Esta tese analisou o processo de difusão e de adoção de uma inovação social gerada em uma organização sem fins lucrativos, com atividades geradoras de renda: o programa A União Faz a Vida do Sistema Sicredi. O método de pesquisa utilizado foi o estudo de caso único com múltiplas unidades de análise: sete municípios gaúchos (Agudo, Arroio do Tigre, Cerro Branco, Candelária, Sobradinho, Paraíso do Sol e Vale do Sol). Foram aplicadas quatro técnicas de coleta de dados: documentação; entrevistas em profundidade com 15 pessoas; observação direta e um grupo focal com 13 participantes. Para a análise do processo de difusão de inovações sociais, uniram-se os pressupostos básicos desenvolvidos por Katz, Levin e Hamilton (1963); por Katz (1999); e por Rogers (de 1962 a 2003) à luz da inovação social. Os principais resultados encontrados apontam algumas especificidades do processo de difusão da inovação social: a ausência da separação entre os processos de difusão e de adoção, visto que eles acontecem de forma interligada; a necessidade de colaboração entre os atores para que a inovação seja aceita; e a presença de uma forte questão política e cultural, elementos capazes de fomentar a adoção ou a rejeição da inovação social. / This thesis analyzed the process of diffusion and the adoption of social innovation generated in a non-profit organization that has income-generating activities. The name of the Sicredi System program is: A União Faz a Vida (Unity is Life). The research method used was a single case study with analysis units in seven cities in the state of Rio Grande do Sul: Agudo, Arroio do Tigre, Cerro Branco, Candelária, Sobradinho, Paraíso do Sol and Vale do Sol). Four data-collection techniques were employed: documentation, in-depth interviews with 15 people, direct observation and a focus group with 13 participants. The basic social innovation suppositions developed by Katz, Levin and Hamilton in1963, by Katz in1999 and by Rogers from 1962 to 2003 were jointly used for the analysis of the diffusion process of social innovations. The main findings identify some specific points in the diffusion process of social innovation: the lack of separation between the processes of diffusion and adoption since they take place simultaneously; the need for collaboration among the different players for innovation to be accepted; and the presence of strong political and cultural elements to promote the adoption or the rejection of social innovation.

Macrophages are the largest cells in the immune system and they regulate inflammatory signalling and inform cell fate decisions. Many signals, including those mediated by Tumor Necrosis Factor alpha (TNFα) converge on a few key intracellular signalling pathways, including the Nuclear Factor kappa B (NFκB) network. The NFκB signalling pathway plays a vital role in the regulation of many different cellular responses, including the production of TNFα itself, which is required to sustain and propagate immune responses to, for example, infection or tissue damage. In this thesis we report on studies-both experimental and theoretical-of the NFκB signalling pathway in macrophages. Our collaborators stimulated these cells with various doses of Lipopolysaccharide (LPS), a molecule that forms the major component of the outer membrane of Gram-negative bacteria: in these experiments it serves as a proxy for bacterial infection. The macrophages, studied in vitro, respond as they are believed to do in tissues, by secreting certain signalling molecules called cytokines: the level of secretion proved to depend on the strength of the LPS stimulus. Further, heterogeneity of macrophage signalling was observed in response to a range of LPS doses. Within individual macrophages LPS stimulation results in oscillatory behaviour of NFκB localisation-NFκB shuttles in and out of the nucleus-with an amplitude (peak nuclear concentration) that also depends on the LPS dose. Heterogeneity was also observed in cells that were stimulated with the same dose intensity. This raises an important question about how immune cells coordinate inflammatory activity in the presence of this variability. In this thesis we aim to achieve an understanding of the system through the qualitative analysis of mathematical models of it. This work explores both the parametric sensitivity and bifurcation analyses for two mathematical models of NFκB in macrophages. Parametric sensitivity analysis is used to investigate the role of parameters on the model's output, especially on certain features of the signal-peak amplitudes, inter-peak intervals and areas beneath curves-that are commonly measured in single-cell experiments. Local bifurcation analysis is conducted in order to show all the possible behaviours produced when varying parameters.

Made available in DSpace on 2015-03-26T13:35:24Z (GMT). No. of bitstreams: 1 texto completo.pdf: 1414332 bytes, checksum: 6e2f42018f3e3dcdf9e8cbccab567e7a (MD5) Previous issue date: 2006-03-20 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Pattern, structure and emergent collective properties are ubiquitous in systems with many units (alive or inanimated) coupled through nonlinear interactions. Within this context, the study of cooperative phenomena in population dynamics of ecological interest has attracted the attention of the mathematicians and physicists since Lotka and Volterra in the 1920s. Thenceforth, in addition to differential equations, theoretical ecology has continuously incorporated powerful and well-established techniques of contacts processes, cellular automata models and others, developed in the fields of condensed matter physics, statistical physics and computational physics. In the present work, a predator-prey model with pursuit and escape triggered by alarm pheromones is proposed and studied through analytical methods and computer simulations. Such models can show oscillatory behavior of the population density, phase transitions that belong to distinct universality classes and rich stationary phase diagrams. Two distinct levels of description were used. In a first approach, we consider a model of cellular automata in which predators and preys walk on a square lattice, according specific rules for each species, in a homogeneous environment and with periodic boundary conditions. The second part of our study is based on the analysis of partial differential equations that also describes the dynamics of a prey-predator system with the same characteristics above. Both, spatially uniform or mean field like and explicit spatio-temporal partial differential equations were considered. These models can represent relevant tools to design better strategies of biological control of pests by predators. In successful cases, the pests and its predators must persist in stable interactions at a low level of pest density. / Padrões, estruturas, propriedades coletivas emergentes são ubíquas em sistemas com muitas unidades (vivas ou inanimadas) acopladas por meio de interações não-lineares. Dentro desse contexto, o estudo de fenômenos cooperativos em dinâmica de populações de interesse ecológico tem atraído a atenção de físicos e matemáticos desde os anos de 1920 com Lotka e Volterra. Portanto, além de equações diferenciais, a teoria ecológica tem continuamente incorporado poderosas e bem-estabelecidas técnicas dos processos de contatos, modelos de autômatos celulares e outros, desenvolvidos no campo de física da matéria condensada, física estatística e física computacional. No presente trabalho, um modelo presa-predador com perseguição e fuga mediada por um feromônio de alarme é proposto e estudado através de métodos analíticos e simulações computacionais. Tais modelos podem exibir comportamentos oscilatórios da densidade de população, transições de fases que pertencem a classes de universalidade distintas e um diagrama de fases rico. Duas abordagens distintas de descrição foram usadas. Numa primeira abordagem, propomos um modelo de Autômato Celular (AC) onde predadores e presas se movimentam, segundo regras específicas para cada espécie, num ambiente homogêneo e com condições de contorno periódicas. A outra parte do nosso estudo é baseado na análise de EDP s que também descrevem a dinâmica de um sistema presa-predador com as mesmas características citadas acima. É feito um estudo considerando as equações sem termos espaciais, isto é, tipo campo médio e depois considerando esses termos. Esses modelos podem representar ferramentas relevantes para o estudo das melhores estratégias para o controle biológico de pragas por predadores. Em casos bem sucedidos, as pestes e seus predadores devem persistir em interações estáveis e com uma baixa densidade da população de pragas.

L’étude des phénomènes de la diffusion d’information à grand échelle est un domaine récent. La diffusion d’information est définie comme le processus de communication par lequel une idée ou une information se propage dans une population et qui peut impacter le comportement des individus. Les institutions, tout comme les entreprises, cherchent à comprendre et à prévoir l’impact de la propagation d’information sur les individus. Une approche de modélisation et simulation permet de mieux comprendre ce processus social et de répondre à ces questions. La modélisation et la simulation à base d’agents offre une approche puissante pour modéliser un tel processus social. Toutefois, les modèles actuels simplifient fortement les facteurs culturels et les informations représentées dans le modèle ainsi que les différents liens interconnectant les individus. Ces éléments sont centraux et déterminants pour le processus de propagation. Afin d'améliorer les modèles de propagation, nous explorons dans cette thèse une représentation de la population plus réaliste. Nous proposons une architecture de modélisation et simulation permettant de simuler les phénomènes de propagation au sein des réseaux sociaux multiplexes et dynamiques basée sur le formalisme DEVS. / The study of information dissemination phenomena of large-scale is a new field. Diffusion of information is defined as the communication process by which an idea or information spreads within a social system and that can impact the behaviour of individuals. Institutions and firms search to understand and predict the impact of information propagation on individuals. Agent-based modelling is a powerful approach for studying such a collective process. However, existing models oversimplify adopters' cultural attributes, different type of link and information content, despite the evidence of their central role in diffusion process. In this thesis, we challenge the feasibility and utility of a more realistic representation of population. We use the DEVS formalism, extended by DS-DEVS and PDEVS in order to simulate the propagation phenomena within a multi-layer social network (MSN).

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-01-12T14:37:07Z No. of bitstreams: 1 eduardochinelatecosta.pdf: 1021822 bytes, checksum: b72dff6cf071e8de1cb23f6cb7d27245 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-01-25T17:15:42Z (GMT) No. of bitstreams: 1 eduardochinelatecosta.pdf: 1021822 bytes, checksum: b72dff6cf071e8de1cb23f6cb7d27245 (MD5) / Made available in DSpace on 2016-01-25T17:15:42Z (GMT). No. of bitstreams: 1 eduardochinelatecosta.pdf: 1021822 bytes, checksum: b72dff6cf071e8de1cb23f6cb7d27245 (MD5) Previous issue date: 2015-02-23 / FAPEMIG - Fundação de Amparo à Pesquisa do Estado de Minas Gerais / Atualmente, há um grande interesse em investigar a dinâmica em Grafos Variantes no Tempo (GVTs). Esses grafos contemplam a evolução temporal, tanto de nós, quanto de arestas. Nesse cenário, de maneira similar a grafos estáticos, o conceito de centralidade geralmente se refere a métricas que avaliam a importância relativa dos vértices. Entretanto, GVTs possibilitam a avaliação da importância dos instantes de tempo (ou estados) de um grafo ao longo de sua existência. Determinar instantes de tempo importantes nesse contexto pode ter aplicações práticas fortes, sendo particularmente úteis para definir melhores momentos para difusão, gerar modelos e prever o comportamento de GVTs. Neste trabalho, nós definimos Centralidade de Tempo em Grafos Variantes no Tempo. A centralidade de tempo avalia a importância relativa dos instantes de tempo. São apresentadas duas métricas de centralidade de tempo voltadas a processos de difusão de informação e uma métrica baseada na disposição das conexões da rede. As métricas foram avaliadas em um conjunto de dados real. Os resultados mostram que os instantes de tempo melhor classificados, de acordo com as métricas criadas, podem tornar o processo de difusão mais rápido e eficiente. Comparado com uma escolha aleatória, o processo de difusão iniciado nos instantes de tempo mais bem classificados pode ser até 2,5 vezes mais rápido, e também pode atingir praticamente o dobro do número de nós na rede em alguns casos. / Currently, there is a great interest in investigating dynamics in Time-Varying Graphs (TVGs). These graphs contemplate the temporal evolution, both nodes and edges. In this scenario, similar to static graphs, centrality usually refers to metrics that assess the relative importance of vertices. However, in TVGs it is possible to assess the importance of time instants (or states) of a graph throughout its existence. Determining important time instants in this context may have strong practical applications and is particularly useful for defining best times to spread, generate models and predict the behavior of TVGs. In this paper, we define time centrality in Time-Varying Graphs. Time centrality evaluates the relative importance of time instants. We present two time centrality metrics focused on information dissemination processes and another based on layout of network connections.. We evaluate metrics we define relying in a real dataset from an hospital environment. Our results show that the best classified time instants, according to created metrics, can make a faster and more efficient diffusion process. Compared to a random choice, the diffusion process starting at best rated time instants can up to 2.5 times faster, and it also can reach almost double the number of nodes in the network in some cases.