Desde os primeiros anos do século XXI mais da metade da população mundial passou a habitar em regiões urbanas. Este processo de urbanização, acelerado a partir da revolução industrial, trouxe consigo benefícios inegáveis à humanidade, mas também possui externalidades evidentes como o congestionamento, a violência e a segregação. Preocupada com estes problemas, surge a ciência do planejamento urbano, ora atuando na antecipação dos fatos, ora remediando os problemas existentes, utilizando as mais variadas propostas que partem de correntes de pensamento compostas por estudiosos das cidades que, nesta difícil tarefa, procuram embasamento em experiências anteriores ou em estudos que invariavelmente envolvem outras áreas de conhecimento, como sociologia, economia e engenharia. Paralelamente a estes esforços, surge nos últimos 40 anos, outra área de conhecimento, denominada Sistemas Complexos, para estudar fenômenos físicos, biológicos, econômicos e sociais, entre outros, utilizando as técnicas provenientes da dinâmica não linear, da física estatística e da computação baseada em agentes, e que vem sendo progressivamente aplicada ao estudo das cidades. Neste trabalho é realizada uma breve revisão da história do surgimento e da evolução das cidades, apresentada na seção 2, e em seguida, na seção 3, é apresentada a teoria dos sistemas complexos, descrevendo as principais características dos sistemas que podem ser analisados à ótica desta área de conhecimento. Na seção 4 é exposto o embasamento teórico e empírico que permite a fusão entre as duas ciências (Planejamento Urbano e Sistemas Complexos). Na seção 5 são reproduzidos alguns estudos computacionais da expansão urbana, consolidados na literatura científica que utilizam técnicas inerentes à ótica de sistemas complexos. Na seção 6 é proposto um modelo computacional simplificado que permite a visualização, extração de dados e análise da ocupação territorial com ênfase no estudo do custo de deslocamento no interior de um centro urbano, permitindo a conclusão, apresentada na seção 7, de que a cidade é um sistema complexo e que a abordagem através da união das áreas de conhecimento é promissora quando trata de assuntos relativos à expansão urbana, podendo ser de grande valor na elaboração de propostas que visem a amenização ou eliminação das externalidades que há tempo têm sido motivo de preocupação para os habitantes dos centros urbanos. / Since the beginning of the XXI century, more than half of the world population lives in urban areas. This process, accelerated with the industrial revolution, brought undeniable advantages, but also several handicaps, externalities such as traffic jams, violence and segregation. Concerned with these problems, a Urban Planning science is developed, acting in the effects as well as in the causes of the problems. In this task, the scientists use tools originated in several fields, such as economy, sociology and engineering. Meanwhile, in the last four decades, a different approach appears. It is called Complex Systems theory, targeted to deal with complex physical, biological, economical and social phenomena, utilizing techniques borrowed from the Non-linear dynamics, from the Statistical physics and from the Agent based models. It has been progressively applied to the Urban Planning science. This work reviews the appearance and development of the cities in the section 2, and in the section 3, the principles of Complex Systems theory are presented. In the section 4 it is showed that the Urban Planning science can profit from the Complex Systems approach. Section 5 reproduces some previous early computational models from the literature. This work contribution is presented next, in the section 6, a simplified model, based on Complex System approach, which allows visualization and urban occupation analyses in function of the transportation costs. The last section concludes that the view of the city as a complex system has many advantages when dealing with the urban expansions and the externalities arising from this process.

Esta dissertação apresenta uma proposta de modelo de mercado financeiro artificial que reproduz séries de retornos com propriedades estatísticas universais semelhantes às observadas em séries reais. Dentre as propriedades, também chamadas de fatos estilizados na Economia, as séries artificiais de retornos exibiram ausência de autocorrelação para os retornos simples, leis de potência para autocorrelação para os retornos absolutos e quadráticos, excesso de curtose nas distribuições de retorno, gaussianidade agregacional e volatilidade clusterizada. Cabe salientar, que não há na literatura um outro mercado artificial que reproduziu tantos fatos estilizados conjuntamente. O modelo dinâmico e síncrono é baseado em agentes que transacionam ativos com risco como ações de empresa através de ordens de compra e venda enviadas ao mercado a cada período de tempo. O preço de mercado das ações é calculado da média ponderada pelo volume das ordens negociadas entre os agentes. O objetivo dos agentes é maximizar sua riqueza e, para isso, seguem ou a estratégia fundamentalista utilizando os dividendos para calcular os preços das ações ou a estratégia técnica baseada em análise de séries temporais. A principal contribuição da modelagem foi acrescentar às estratégias um fator de aprendizado em que o agente considera sua habilidade individual passada de previsão de riqueza esperada para calcular os retornos futuros. Este trabalho também mediu o coeficiente de Gini para descobrir como algumas variáveis de mercado afetavam a distribuição de riqueza dos agentes e, além disso, estudou quais valores de dividendo tornavam uma estratégia mais eficiente que outra. Por fim, incorporaram-se características evolutivas aos agentes possibilitandoos a trocar de estratégias no decorrer da simulação e, com isso, os resultados mostraram aumento da riqueza dos agentes. / This work presents a new artificial stock market model for reproducing price time series of assets in such market model. For a suitable validation of the model, we verified several statistical and universal properties (called stylized facts in the Economics Literature) and similar results are obtained with data extracted from real stock markets. We investigate several properties including absence of autocorrelation for simple returns and the power behavior law of autocorrelation for absolute and quadratic returns, excess of kurtosis, aggregational gaussianity, and clustered volatility. It is important to mention that no other similar artificial model has investigated so many statistical universalities. Our synchronous model is based on agents negotiating risk assets through purchase and sale orders. These orders are stored in books for each simulation step. The weighted average volume of all orders negotiated by the agents determines the price of an asset. For the sake of simplicity, our model considers two kinds of strategies: 1. Fundamentalist - where one uses the dividends to calculate the expected return of an asset; 2. Trend predictor - where one obtains the expected returns directly from an analysis of the price time series. One of the main contributions of our model was to add a term that works as the expected wealth of an agent. This is considered an important psychological factor in the decision making process. In addition, we consider an income inequality index to analyze the wealth distribution of the agents: the Gini-coefficient, which predicts an inequality interval of [0 (society completely fair),1 (society completely unfair)]. We also study the influence of the dividends and risk free assets parameters on this coefficient. Finally, some evolutionary features of the model are analyzed. Our results show an increase in agent’s wealth when strategies are updated according to the following criteria: if expected wealth does not reach a given threshold, the agent changes his strategy from Fundamentalist to Trend Predictor or vice-versa. If the expected wealth reaches the specified threshold, the agent keeps his initial strategy. We tested different threshold values in this analysis and the conclusion was confirmed in all cases studied.

Since the wave of civil violence in the USA in the 1960s, many social theorists have tried to explain why riots occur. Despite at least 50 years of research since then, there is still not enough insight to anticipate large events like the 2011 Arab Spring and London riots. The main goal of this thesis is therefore to improve understanding about how underlying conditions influence and drive riot dynamics, such as the intensity, spread, and duration. I develop a new mechanistic and stochastic agent-based model for riots. Previous models have either only targeted general phenomena associated with riots, or aimed at behaviour specific to a single event. In this thesis I combine both approaches: I demonstrate how the model in which the motivation of the agents is based on general concepts, can be applied to the specific situation of the 2011 London riots. The model reproduces the majority of the behaviour observed in the London riots (r = 0.4-0.8). One of the key factors under investigation is the relationship between protests and outbursts of civil violence. Riots are often preceded by protests, such that a large pool of potential rioters is directly available. I find that the number of times a protest is repeated has greater influence on riot dynamics than the protest crowd size. The support shown during demonstrations might incite false confidence in individuals, potentially leading to quicker escalation. Another question is how contact networks and collective identity influence the spread of violence between different locations. The role of online social media (e.g. Twitter) has been a major focus in trying to explain why the violence in the 2011 Arab spring spread so quickly and so far. I investigate the role of social similarity as another factor that might have contributed to the diffusion of unrest, and demonstrate the existence of a critical transition in riot activity when increasing the density of the contact network in the model. Such increases in density beyond the critical thresholds might have been introduced by online social networks. Finally, I explore the sensitivity to cooperation of different potential riot groups. In some cases, mixed populations with different collective identities can form coalitions within neighbourhoods based on shared grievances, which could lead to increases in riot size and riot probability. I examine the influence of the social structure and spread of these populations over different neighbourhoods, as well as the overlap in grievances and different demographic structures.

Esta dissertação apresenta uma proposta de modelo de mercado financeiro artificial que reproduz séries de retornos com propriedades estatísticas universais semelhantes às observadas em séries reais. Dentre as propriedades, também chamadas de fatos estilizados na Economia, as séries artificiais de retornos exibiram ausência de autocorrelação para os retornos simples, leis de potência para autocorrelação para os retornos absolutos e quadráticos, excesso de curtose nas distribuições de retorno, gaussianidade agregacional e volatilidade clusterizada. Cabe salientar, que não há na literatura um outro mercado artificial que reproduziu tantos fatos estilizados conjuntamente. O modelo dinâmico e síncrono é baseado em agentes que transacionam ativos com risco como ações de empresa através de ordens de compra e venda enviadas ao mercado a cada período de tempo. O preço de mercado das ações é calculado da média ponderada pelo volume das ordens negociadas entre os agentes. O objetivo dos agentes é maximizar sua riqueza e, para isso, seguem ou a estratégia fundamentalista utilizando os dividendos para calcular os preços das ações ou a estratégia técnica baseada em análise de séries temporais. A principal contribuição da modelagem foi acrescentar às estratégias um fator de aprendizado em que o agente considera sua habilidade individual passada de previsão de riqueza esperada para calcular os retornos futuros. Este trabalho também mediu o coeficiente de Gini para descobrir como algumas variáveis de mercado afetavam a distribuição de riqueza dos agentes e, além disso, estudou quais valores de dividendo tornavam uma estratégia mais eficiente que outra. Por fim, incorporaram-se características evolutivas aos agentes possibilitandoos a trocar de estratégias no decorrer da simulação e, com isso, os resultados mostraram aumento da riqueza dos agentes. / This work presents a new artificial stock market model for reproducing price time series of assets in such market model. For a suitable validation of the model, we verified several statistical and universal properties (called stylized facts in the Economics Literature) and similar results are obtained with data extracted from real stock markets. We investigate several properties including absence of autocorrelation for simple returns and the power behavior law of autocorrelation for absolute and quadratic returns, excess of kurtosis, aggregational gaussianity, and clustered volatility. It is important to mention that no other similar artificial model has investigated so many statistical universalities. Our synchronous model is based on agents negotiating risk assets through purchase and sale orders. These orders are stored in books for each simulation step. The weighted average volume of all orders negotiated by the agents determines the price of an asset. For the sake of simplicity, our model considers two kinds of strategies: 1. Fundamentalist - where one uses the dividends to calculate the expected return of an asset; 2. Trend predictor - where one obtains the expected returns directly from an analysis of the price time series. One of the main contributions of our model was to add a term that works as the expected wealth of an agent. This is considered an important psychological factor in the decision making process. In addition, we consider an income inequality index to analyze the wealth distribution of the agents: the Gini-coefficient, which predicts an inequality interval of [0 (society completely fair),1 (society completely unfair)]. We also study the influence of the dividends and risk free assets parameters on this coefficient. Finally, some evolutionary features of the model are analyzed. Our results show an increase in agent’s wealth when strategies are updated according to the following criteria: if expected wealth does not reach a given threshold, the agent changes his strategy from Fundamentalist to Trend Predictor or vice-versa. If the expected wealth reaches the specified threshold, the agent keeps his initial strategy. We tested different threshold values in this analysis and the conclusion was confirmed in all cases studied.

Submitted by Henrique Teixeira (henrique.oliv@gmail.com) on 2016-03-16T02:51:18Z No. of bitstreams: 1 Dissertacao.pdf: 1894861 bytes, checksum: af73e440cc555c69c32dbb74b4ba3f59 (MD5) / Approved for entry into archive by Renata de Souza Nascimento (renata.souza@fgv.br) on 2016-03-16T21:49:10Z (GMT) No. of bitstreams: 1 Dissertacao.pdf: 1894861 bytes, checksum: af73e440cc555c69c32dbb74b4ba3f59 (MD5) / Made available in DSpace on 2016-03-17T11:42:41Z (GMT). No. of bitstreams: 1 Dissertacao.pdf: 1894861 bytes, checksum: af73e440cc555c69c32dbb74b4ba3f59 (MD5) Previous issue date: 2016-02-18 / Starting from the idea that economic systems fall into complexity theory, where its many agents interact with each other without a central control and that these interactions are able to change the future behavior of the agents and the entire system, similar to a chaotic system we increase the model of Russo et al. (2014) to carry out three experiments focusing on the interaction between Banks and Firms in an artificial economy. The first experiment is relative to Relationship Banking where, according to the literature, the interaction over time between Banks and Firms are able to produce mutual benefits, mainly due to reduction of the information asymmetry between them. The following experiment is related to information heterogeneity in the credit market, where the larger the bank, the higher their visibility in the credit market, increasing the number of consult for new loans. Finally, the third experiment is about the effects on the credit market of the heterogeneity of prices that Firms faces in the goods market. / Partindo da ideia de que os sistemas econômicos se enquadram na teoria da complexidade, onde seus inúmeros agentes interagem entre si sem um controle central e que essas interações são capazes de alterar o comportamento futuro dos agentes e de todo o sistema, semelhante a um sistema caótico, incrementamos o modelo de Russo et al. (2014) para a realização de três experimentos com foco na interação entre bancos e empresas em uma economia artificial. O primeiro experimento diz respeito a Relationship Banking onde, segundo a literatura, a interação ao longo do tempo entre bancos e empresas é capaz de produzir benefícios mútuos, principalmente devido a redução da assimetria de informação entre eles. O experimento seguinte está relacionado a assimetria de informação no mercado de crédito, onde quanto maior o banco, maior sua visibilidade no mercado de crédito, elevando na mesma proporção as consultar para novos emprestimos. Por fim, o terceiro experimento é relativo aos efeitos no mercado de crédito da heterogeneidade de preços que as empresas se deparam no mercado de bens

Esta dissertação apresenta uma proposta de modelo de mercado financeiro artificial que reproduz séries de retornos com propriedades estatísticas universais semelhantes às observadas em séries reais. Dentre as propriedades, também chamadas de fatos estilizados na Economia, as séries artificiais de retornos exibiram ausência de autocorrelação para os retornos simples, leis de potência para autocorrelação para os retornos absolutos e quadráticos, excesso de curtose nas distribuições de retorno, gaussianidade agregacional e volatilidade clusterizada. Cabe salientar, que não há na literatura um outro mercado artificial que reproduziu tantos fatos estilizados conjuntamente. O modelo dinâmico e síncrono é baseado em agentes que transacionam ativos com risco como ações de empresa através de ordens de compra e venda enviadas ao mercado a cada período de tempo. O preço de mercado das ações é calculado da média ponderada pelo volume das ordens negociadas entre os agentes. O objetivo dos agentes é maximizar sua riqueza e, para isso, seguem ou a estratégia fundamentalista utilizando os dividendos para calcular os preços das ações ou a estratégia técnica baseada em análise de séries temporais. A principal contribuição da modelagem foi acrescentar às estratégias um fator de aprendizado em que o agente considera sua habilidade individual passada de previsão de riqueza esperada para calcular os retornos futuros. Este trabalho também mediu o coeficiente de Gini para descobrir como algumas variáveis de mercado afetavam a distribuição de riqueza dos agentes e, além disso, estudou quais valores de dividendo tornavam uma estratégia mais eficiente que outra. Por fim, incorporaram-se características evolutivas aos agentes possibilitandoos a trocar de estratégias no decorrer da simulação e, com isso, os resultados mostraram aumento da riqueza dos agentes. / This work presents a new artificial stock market model for reproducing price time series of assets in such market model. For a suitable validation of the model, we verified several statistical and universal properties (called stylized facts in the Economics Literature) and similar results are obtained with data extracted from real stock markets. We investigate several properties including absence of autocorrelation for simple returns and the power behavior law of autocorrelation for absolute and quadratic returns, excess of kurtosis, aggregational gaussianity, and clustered volatility. It is important to mention that no other similar artificial model has investigated so many statistical universalities. Our synchronous model is based on agents negotiating risk assets through purchase and sale orders. These orders are stored in books for each simulation step. The weighted average volume of all orders negotiated by the agents determines the price of an asset. For the sake of simplicity, our model considers two kinds of strategies: 1. Fundamentalist - where one uses the dividends to calculate the expected return of an asset; 2. Trend predictor - where one obtains the expected returns directly from an analysis of the price time series. One of the main contributions of our model was to add a term that works as the expected wealth of an agent. This is considered an important psychological factor in the decision making process. In addition, we consider an income inequality index to analyze the wealth distribution of the agents: the Gini-coefficient, which predicts an inequality interval of [0 (society completely fair),1 (society completely unfair)]. We also study the influence of the dividends and risk free assets parameters on this coefficient. Finally, some evolutionary features of the model are analyzed. Our results show an increase in agent’s wealth when strategies are updated according to the following criteria: if expected wealth does not reach a given threshold, the agent changes his strategy from Fundamentalist to Trend Predictor or vice-versa. If the expected wealth reaches the specified threshold, the agent keeps his initial strategy. We tested different threshold values in this analysis and the conclusion was confirmed in all cases studied.

In the thesis we focus on involving Agent-based models in modeling financial markets. In Agent-based models of economical systems, often called Agent-based computational economics (ACE), market price is established by actions and interactions of autonomous agents using heuristics or simple decision-making rules. This approach to modeling of financial markets provide us with better understanding of establishing market price then aggregate economical models which focuses exclusively on societally "optimal" equilibria assuming that they are achieved by informed and rational behavior of people. The thesis consists of two main parts. The first one, theoretical, is an introduction to agent based modeling, bounded rationality and social network Our concern in the second part of the thesis is a model of volatility on financial markets. This model is interesting example of agent based approach to creating economical models. However it contains some non-realistic assumption from which the most controversial is the space where agents interacts -- two dimensional lattice. In this part of the work the model is converted from two dimensional lattice to the networks which better corresponds to real social networks but we also experiment with another extension of the agent's decision-making function. The intended outcome of the work is verifying the quality of the model, to learn the effect of our model extensions on price volatility, overview of attributes of the particular networks and discussion whether such models could provide some valuable information to the economist which are interested in financial markets.

We consider Sequential Monte Carlo Approximate Bayesian Computation (SMC ABC) as a method of calibration for the use of agent based models in market micro-structure. To date, there are no successful calibrations of agent based models to high frequency trading data. Here we test whether a more sophisticated calibration technique, SMC ABC, will achieve this feat on one of the leading agent based models in high frequency trading literature (the Preis-Golke-Paul-Schneider Agent Based Model (Preis et al., 2006)). We find that, although SMC ABC's naive approach of updating distributions can successfully calibrate simple toy models, such as autoregressive moving average models, it fails to calibrate this agent based model for high frequency trading. This may be for two key reasons, either the parameters of the model are not uniquely identifiable given the model output or the SMC ABC rejection mechanism results in information loss rendering parameters unidentifiable given insucient summary statistics.

Infectious diseases caused by bacteria, fungi, viruses and parasites have affected humans historically. Infectious diseases remain a major cause of premature death and a public health concern globally with increased mortality and significant economic burden. Unvaccinated individuals, people with suppressed and compromised immune systems are at higher risk of suffering from infectious diseases. In spite of significant advancements in infectious diseases research, the control or treatment process faces challenges. The mucosal immune system plays a crucial role in safeguarding the body from harmful pathogens, while being constantly exposed to the environment. To develop treatment options for infectious diseases, it is vital to understand the immune responses that occur during infection. The two infectious diseases presented here are: i) Helicobacter pylori infection and ii) human immunodeficiency (HIV) and human papillomavirus (HPV) co-infection. H pylori, is a bacterium that colonizes the stomach and causes gastric cancer in 1-2% but is beneficial for protection against allergies and gastroesophageal diseases. An estimated 85% of H pylori colonized individuals show no detrimental effects. HIV is a virus that causes AIDS, one of the deadliest and most persistent epidemics. HIV-infected patients are at an increased risk of co-infection with HPV, and report an increased incidence of oral cancer. The goal of this thesis is to elucidate the host immune responses in infectious diseases via the use of computational and mathematical models. First, the thesis reviews the need for computational and mathematical models to study the immune responses in the course of infectious diseases. Second, it presents a novel sensitivity analysis method that identifies important parameters in a hybrid (agent-based/equation-based) model of H. pylori infection. Third, it introduces a novel model representing the HIV/HPV coinfection and compares the simulation results with a clinical study. Fourth, it discusses the need of advanced modeling technologies to achieve a personalized systems wide approach and the challenges that can be encountered in the process. Taken together, the work in this dissertation presents modeling approaches that could lead to the identification of host immune factors in infectious diseases in a predictive and more resource-efficient manner. / Doctor of Philosophy / Infectious diseases caused by bacteria, fungi, viruses and parasites have affected humans historically. Infectious diseases remain a major cause of premature death and a public health concern globally with increased mortality and significant economic burden. These infections can occur either via air, travel to at-risk places, direct person-to-person contact with an infected individual or through water or fecal route. Unvaccinated individuals, individuals with suppressed and compromised immune system such as that in HIV carriers are at higher risk of getting infectious diseases. In spite of significant advancements in infectious diseases research, the control and treatment of these diseases faces numerous challenges. The mucosal immune system plays a crucial role in safeguarding the body from harmful pathogens, while being exposed to the environment, mainly food antigens. To develop treatment options for infectious diseases, it is vital to understand the immune responses that occur during infection. In this work, we focus on gut immune system that acts like an ecosystem comprising of trillions of interacting cells and molecules, including membars of the microbiome. The goal of this dissertation is to develop computational models that can simulate host immune responses in two infectious diseases- i) Helicobacter pylori infection and ii) human immunodeficiency virus (HIV)-human papilloma virus (HPV) co-infection. Firstly, it reviews the various mathematical techniques and systems biology based methods. Second, it introduces a "hybrid" model that combines different mathematical and statistical approaches to study H. pylori infection. Third, it highlights the development of a novel HIV/HPV coinfection model and compares the results from a clinical trial study. Fourth, it discusses the challenges that can be encountered in adapting machine learning based computational technologies. Taken together, the work in this dissertation presents modeling approaches that could lead to the identification of host immune factors in infectious diseases in a predictive and more resourceful way.

Approved for public release; distribution is unlimited / The US Department of Defense is currently pursuing the most comprehensive transformation of its forces since the early years of WWII. This transformation is a holistic approach to update both the equipment that the forces will fight its conflicts with and the way in which they will fight. This transformation relies heavily on fully networked air, ground and space based platforms. While many experts agree that in the course of the next 10 years communications equipment will emerge to support the networking of these systems, there remains much uncertainty on how operations will be effected if the technology does not mature enough to meet expectations. This research shows that even a 25 percent degradation in communications range could pose significant challenges for this Future Force. Additionally, even small delays (latencies greater than one minute) and constraints on network throughput can increase the Future Force casualties and the duration of battle. While the end result in all analysis shows that the Future Force is a superior element with the same battle end state-victory, the cost of that victory depends significantly on effective communications. / Captain, United States Army