[pt] Esta dissertação apresenta um estudo sobre a Logística de Distribuição de Derivados de Petróleo, com o desenvolvimento de um modelo matemático para a determinação do custo logístico de distribuição de estoques estratégicos de combustíveis no Brasil. Este modelo não se propõe a determinar o custo ótimo de distribuição e sim estabelecer uma distribuição racional dos estoques estratégicos de combustíveis, com base na malha real de distribuição. Para o desenvolvimento do modelo, primeiramente foi realizado um diagnóstico da distribuição nacional de derivados energéticos de petróleo, mais especificamente gasolina, diesel, querosene de aviação, óleo combustível e gás liquefeito de petróleo. Para cada um desses produtos determinou-se a malha de distribuição, com os fluxos de produtos entre refinarias e terminais até as bases de combustíveis secundárias, com seus respectivos modais. Em uma segunda etapa foi construída uma base de dados georreferenciada no software TransCAD, com os dados levantados e diagnosticados anteriormente. Este sistema de informação geográfica serviu como suporte para análises logísticas e como facilitador na visualização e entendimento dos resultados. Os resultados dos custos logísticos de distribuição são mostrados por produto, para evidenciar o caráter diferenciado da distribuição de cada derivado. / [en] This master thesis presents a study about the petroleum products distribution system, with the development of a mathematical model for the determination of the logistic costs of strategic oil stocks in Brazil. This model doesn't intend to determine the minimum distribution cost , but establishes a rational distribution of the strategic oil stocks, based in the real distribution network. For the development of the model, firstly a diagnosis of the national distribution was accomplished, more specifically gasoline, diesel, aviation kerosene, combustible oil and liquefied gas of petroleum. For each one of those products, the distribution network was determined, with the flows of products among refineries and terminals until the secondary bases of fuels, with its respective transportation modals. In a second stage, a geographical database was built in the software TransCAD, with all available data . This geographical information system was a support tool for logistic analyses, visualization and better understanding of the results. The results of the total distribution costs are presented by product, to evidence the differentiated character of the distribution of each product.

In the present paper, we identify several distributions from Physics and study their applicability to phenomena such as distribution of income, wealth, and expenditure. Firstly, we apply logistic distribution to these data and we find that it fits very well the annual data for the entire income interval including for upper income segment of population. Secondly, we apply Fermi-Dirac distribution to these data. We seek to explain possible correlations and analogies between economic systems and statistical thermodynamics systems. We try to explain their behaviour and properties when we correlate physical variables with macroeconomic aggregates and indicators. Then we draw some analogies between parameters of the Fermi-Dirac distribution and macroeconomic variables. Thirdly, as complex systems are modelled using polynomial distributions, we apply polynomials to the annual sets of data and we find that it fits very well also the entire income interval. Fourthly, we develop a new methodology to approach dynamically the income, wealth, and expenditure distribution similarly with dynamical complex systems. This methodology was applied to different time intervals consisting of consecutive years up to 35 years. Finally, we develop a mathematical model based on a Hamiltonian that maximises utility function applied to Ramsey model using Fermi-Dirac and polynomial utility functions. We find some theoretical connections with time preference theory. We apply these distributions to a large pool of data from countries with different levels of development, using different methods for calculation of income, wealth, and expenditure.

Os dados de sobrevivência possuem peculiaridades que necessitam de uma atenção especial no momento em que se deseja realizar uma análise nos mesmos. Em tais dados é comum a presença de censuras e sua variável resposta é definida como o tempo de vida até a ocorrência de um evento de interesse. Existem distribuições que acolhem dados de sobrevivência, como as distribuições exponencial, Weibull, gama, gama generalizada, entre outras, assim como seus respectivos modelos de regressão adaptados para esse tipo de estudo. Os modelos de regressão exponencial e Weibull são os mais citados na literatura por terem fácil aplicação e se modelarem bem aos dados. O modelo de regressão gama generalizado geralmente se adapta melhor aos dados por ter três parâmetros, assim como o modelo de regressão log-logístico, que é visto como uma alternativa à distribuição Weibull e é muito utilizado por ter formas explícitas para a sua função de sobrevivência e de falha. No entanto, esses modelos ainda possuem restrições e, por conta disso, novas famílias de modelos de regressão estão sendo desenvolvidas na literatura, assim como a família de distribuições odd log-logística generalizada, que pretende oferecer melhores ajustes pois aparenta ter capacidade de modelar diferentes tipos de dados. O objetivo dessa dissertação foi aplicar técnicas de análise de sobrevivência na modelagem dos tempos de vida de Micrurus corallinus, ajustando os modelos já presentes na literatura e o modelo proposto odd log-logística generalizada Weibull (OLLG-W). Conclui-se que o modelo de regressão que se mostrou adequado aos dados foi o log-logístico e o modelo de regressão OLLG-W não apresentou nenhuma vantagem em relação aos que já são frequentes na literatura. / Survival data hold special attention-needed peculiarities the moment you intend to realize an analysis on. These data own censorships and their variable responses are defined as lifetime to interest- event occurrence. There are distributions that harbor these data, such as exponential distribution, Weibull, gamma, generalized gamma, among others, just as their respective event-adapted regression models. Exponential regression and Weibull models are the most literature recurrent, in view of their easy application and appropriate data modeling. The generalized gamma regression model usually is a better fit to the data, due to its three-parameter comprise, just as the log-logistic regression model, which is seen as an alternative to Weibull distribution and is heavily utilized for it\'s explicit shapes to survivability and fail functions. Nonetheless, these models still retain restrictions and, on account of that, new regression model families are being developed, as in the log logistic generalized distribution family, which intends to offer better settings due to its different real data modeling ability. The purpose of this dissertation was to apply survival analysis techniques in Micrurus corallinus lifetime modeling, adjusting already existing models and the proposed Weibull generalized odd log logistic model (OLLG-W). We came to the conclusion that the adequate regression model to Micrurus corallinus data was the log-logistic model. The OLLG-W model didn\'t offer any benefits when compared to literature-recurrent ones.

Submitted by Aelson Maciera (aelsoncm@terra.com.br) on 2017-09-27T18:09:58Z No. of bitstreams: 1 TeseRFP.pdf: 2142415 bytes, checksum: 8dcd8615da0b442e9f1b52f35364715b (MD5) / Approved for entry into archive by Ronildo Prado (producaointelectual.bco@ufscar.br) on 2017-10-10T18:16:14Z (GMT) No. of bitstreams: 1 TeseRFP.pdf: 2142415 bytes, checksum: 8dcd8615da0b442e9f1b52f35364715b (MD5) / Approved for entry into archive by Ronildo Prado (producaointelectual.bco@ufscar.br) on 2017-10-10T18:16:22Z (GMT) No. of bitstreams: 1 TeseRFP.pdf: 2142415 bytes, checksum: 8dcd8615da0b442e9f1b52f35364715b (MD5) / Made available in DSpace on 2017-10-10T18:23:04Z (GMT). No. of bitstreams: 1 TeseRFP.pdf: 2142415 bytes, checksum: 8dcd8615da0b442e9f1b52f35364715b (MD5) Previous issue date: 2017-08-25 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / The Beta distribution is a bounded domain distribution which has dominated the modeling the distribution of random variable that assume value between 0 and 1. Bounded domain distributions arising in various situations such as rates, proportions and index. Motivated by an analysis of electoral votes percentages (where a distribution with support on the positive real numbers was used, although a distribution with limited support could be more suitable) we focus on alternative distributions to Beta distribution with emphasis in regression models. In this work, initially we present the Simplex mixture model as a flexible model to modeling the distribution of bounded random variable then we extend the model to the context of regression models with the inclusion of covariates. The parameters estimation is discussed for both models considering Bayesian inference. We apply these models to simulated data sets in order to investigate the performance of the estimators. The results obtained were satisfactory for all the cases investigated. Finally, we introduce a parameterization of the L-Logistic distribution to be used in the context of regression models and we extend it to a mixture of mixed models. / A distribuição beta é uma distribuição com suporte limitado que tem dominado a modelagem de variáveis aleatórias que assumem valores entre 0 e 1. Distribuições com suporte limitado surgem em várias situações como em taxas, proporções e índices. Motivados por uma análise de porcentagens de votos eleitorais, em que foi assumida uma distribuição com suporte nos números reais positivos quando uma distribuição com suporte limitado seira mais apropriada, focamos em modelos alternativos a distribuição beta com enfase em modelos de regressão. Neste trabalho, apresentamos, inicialmente, um modelo de mistura de distribuições Simplex como um modelo flexível para modelar a distribuição de variáveis aleatórias que assumem valores em um intervalo limitado, em seguida estendemos o modelo para o contexto de modelos de regressão com a inclusão de covariáveis. A estimação dos parâmetros foi discutida para ambos os modelos, considerando o método bayesiano. Aplicamos os dois modelos a dados simulados para investigarmos a performance dos estimadores usados. Os resultados obtidos foram satisfatórios para todos os casos investigados. Finalmente, introduzimos a distribuição L-Logistica no contexto de modelos de regressão e posteriormente estendemos este modelo para o contexto de misturas de modelos de regressão mista.

Orientador: Antonio Padilha Feltrin / Resumo: Neste trabalho é apresentado um método que permite determinar o crescimento espaço-temporal da demanda de energia elétrica devido às mudanças nos hábitos de consumo no setor residencial. A proposta é baseada em uma regressão ponderada geograficamente que permite determinar a localização espacial dos setores com maior proporção de residências candidatas para comprar um novo eletrodoméstico, e uma regressão de distribuição logística que permite simular em cada setor, como vai ser o crescimento ao longo do tempo dessa proporção de residências candidatas para comprar o aparelho. Finalmente, o método determina o impacto nas curvas de carga dos transformadores de distribuição, considerando: o número de residências candidatas em cada setor, e informações do eletrodoméstico em estudo, tais como: curva de carga em p.u, potência nominal, fator de utilização, fator de coincidência e fator de potência. A região em estudo é dividida em pequenas subáreas, com o objetivo de melhorar a resolução espacial do prognóstico, e também considerar interrelações de proximidade entre as subáreas, para determinar como as decisões tomadas em um local influenciam nas preferências de seus vizinhos. O método proposto usa como dados de entrada variáveis socioeconômicas do censo da população que são de fácil acesso para as empresas do setor elétrico e que caracterizam a economia e as preferências da população da cidade em estudo. O método proposto foi aplicado em uma cidade de médio porte da República do Equ... (Resumo completo, clicar acesso eletrônico abaixo) / Mestre

Submitted by MARIO ANDRES MEJIA ALZATE (marioandretty_17@hotmail.com) on 2017-01-15T06:04:36Z No. of bitstreams: 1 Dissertação Mestrado_final_1 (2).pdf: 10151663 bytes, checksum: 19b32f17aadb3d9188da99327b13cc74 (MD5) / Approved for entry into archive by LUIZA DE MENEZES ROMANETTO (luizamenezes@reitoria.unesp.br) on 2017-01-19T17:23:34Z (GMT) No. of bitstreams: 1 mejiaalzate_ma_me_ilha.pdf: 10151663 bytes, checksum: 19b32f17aadb3d9188da99327b13cc74 (MD5) / Made available in DSpace on 2017-01-19T17:23:34Z (GMT). No. of bitstreams: 1 mejiaalzate_ma_me_ilha.pdf: 10151663 bytes, checksum: 19b32f17aadb3d9188da99327b13cc74 (MD5) Previous issue date: 2016-12-19 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Neste trabalho é apresentado um método que permite determinar o crescimento espaço-temporal da demanda de energia elétrica devido às mudanças nos hábitos de consumo no setor residencial. A proposta é baseada em uma regressão ponderada geograficamente que permite determinar a localização espacial dos setores com maior proporção de residências candidatas para comprar um novo eletrodoméstico, e uma regressão de distribuição logística que permite simular em cada setor, como vai ser o crescimento ao longo do tempo dessa proporção de residências candidatas para comprar o aparelho. Finalmente, o método determina o impacto nas curvas de carga dos transformadores de distribuição, considerando: o número de residências candidatas em cada setor, e informações do eletrodoméstico em estudo, tais como: curva de carga em p.u, potência nominal, fator de utilização, fator de coincidência e fator de potência. A região em estudo é dividida em pequenas subáreas, com o objetivo de melhorar a resolução espacial do prognóstico, e também considerar interrelações de proximidade entre as subáreas, para determinar como as decisões tomadas em um local influenciam nas preferências de seus vizinhos. O método proposto usa como dados de entrada variáveis socioeconômicas do censo da população que são de fácil acesso para as empresas do setor elétrico e que caracterizam a economia e as preferências da população da cidade em estudo. O método proposto foi aplicado em uma cidade de médio porte da República do Equador a fim de determinar o crescimento espaço-temporal da demanda de energia devido à compra de fogões de indução. Os resultados obtidos são mapas que permitem identificar os setores mais vulneráveis para apresentar crescimento da demanda devido à compra do eletrodoméstico. Também são apresentados gráficos que mostram o impacto nas curvas de carga dos transformadores durante o período de estudo estabelecido. Esses resultados fornecem informações importantes que servem de referência no planejamento do sistema de distribuição e do mercado de energia elétrica. / This work presents a method to determine the spatial-temporal growth of electric energy demand due to changes in consumption habits in the residential sector. The proposal is based on a geographically weighted regression that allows us to determine the spatial location of the sectors with the highest proportion of candidate households to buy a new appliance, and a logistic distribution regression that allows us to simulate in each of these sectors, the growth over time, the proportion of households that are candidates to buy this appliance. Finally, the method determines the impact on the load curves of the distribution transformers, considering: the number of candidate households in each sector, and information of the home appliance, such as: load curve in pu, nominal power, utilization factor, Coincidence factor and power factor. The study area is divided into small subareas with the aim of improving the spatial resolution of the prognosis and also considers the interrelation of proximity between the subareas to determine how decisions made in one place can influence the preferences of its neighbors. The input data of the proposed method are socioeconomic variables of the population census, which are easily accessible to companies in the electricity sector, and which characterize the economy and the preferences of the population of the studied city. The method was applied in a medium-sized city of the Republic of Ecuador in order to determine the spatial-temporal growth of energy demand due to the purchase of induction stoves. The results obtained are maps that allow identifying the most vulnerable sectors to show increased demand due to the purchase of the appliance. Also, graphs were obtained that show the impact on the load curves of the transformers during the established study period. These results provide important information that serve as a reference in planning the distribution system and the electricity market.

This thesis aims to investigate the potential effects on Value at Risk (VaR) measurements when including social media sentiments from Reddit and Twitter. The investigated stock companies are Apple, Alphabet and Tesla. Furthermore, the VaR measurements will be computed through volatility forecasts and assumptions about the return distributions. The volatility will be forecasted by two different models and each model will both include and exclude social media sentiments, so there will be four different volatility forecasts for each stock. Moreover, the volatility models will be the Heterogeneous autoregression (HAR) model and the Heterogeneous autoregression Neural Network (HAR-NN) model. The assumptions of return distributions are a log-logistic distribution and a log-normal distribution. In addition to this, the VaR measurements are computed and evaluated through number of breaches for each of the volatility forecasts and for both assumptions of a return distribution. The result shows that there is an improvement in forecasting volatility for Apple and Alphabet, as well as fewer VaR breaches for both assumptions of log-return distributions. However, the results for Tesla showed that the volatility forecasts were better when excluding social media sentiment. A possible reason for this might be due to Twitter posts made by influential people, like Elon Musk that would have a larger effect on the volatility than the average sentiment score over that day. Another possible explanation to this might be due to multicollinearity. Overall, the results showed that the assumption of a log-logistic distribution was more suitable over a log- normal return distribution for all three stocks. / Den här studien undersöker de potentiella effekterna av att inkludera sentiment från Reddit och Twitter vid beräkning av det finansiella riskmåttet VaR. De undersökta aktierna är Apple, Alphabet och Tesla. VaR måtten beräknas genom att förutspå volatiliteten samt genom att göra antaganden om aktiernas avkast- ningsfördelning. Volatiliteten förutspås genom två olika modeller och bägge modeller kommer både att inkludera sentiment från sociala medier samt exkludera sentimenten. Därav kommer det totalt vara fyra olika volatilitets prognoser för vardera aktie. Volatilitetsmodellerna som används i denna studie är HAR modellen och HAR-NN modellen. De antaganden som görs om logartimen av avkastningsfördelningarna är att de följer en logistik fördelning samt en normalfördelning. Dessutom är VaR måtten beräknade och eval- uerade genom antalet gånger portföljen överskrider VaR måttet för varje volatilitetsprognos och för vardera antagande om avkastningsfördelning. Resultaten av denna studie visar att inkludering av sentiment från sociala medier förbättrar volatilitetsprognosen för Apple och Alphabet, samt att portföljen överskrider dessa VaR mått för båda fördelningsantaganden. Däremot, visar resultaten för Tesla att volatilitetsprog- nosen är sämre då sentiment från sociala medier inkluderas i modellerna. En möjlig anledning till detta skulle kunna vara på grund av inflyelserika personer, så som Elon Musk vars Twitter inlägg har större påverkan på aktievolatiliteten än medelsentimentet. En annan anledning till detta skulle kunna vara på grund av multikollinearitet, ifall sentimenten till Tesla är starkt korrelerade med volatiliteten. Samman- taget visade resultaten att antagandet av att logaritmen av avkastningarna följer en logistikt fördelning var mer passande än antagandet av en normalfördelning för alla tre aktier.

Cure rate models, introduced by Boag (1949), are very commonly used while modelling lifetime data involving long time survivors. Applications of cure rate models can be seen in biomedical science, industrial reliability, finance, manufacturing, demography and criminology. In this thesis, cure rate models are discussed under a competing cause scenario, with the assumption of proportional odds (PO) lifetime distributions for the susceptibles, and statistical inferential methods are then developed based on right-censored data. In Chapter 2, a flexible cure rate model is discussed by assuming the number of competing causes for the event of interest following the Conway-Maxwell (COM) Poisson distribution, and their corresponding lifetimes of non-cured or susceptible individuals can be described by PO model. This provides a natural extension of the work of Gu et al. (2011) who had considered a geometric number of competing causes. Under right censoring, maximum likelihood estimators (MLEs) are obtained by the use of expectation-maximization (EM) algorithm. An extensive Monte Carlo simulation study is carried out for various scenarios, and model discrimination between some well-known cure models like geometric, Poisson and Bernoulli is also examined. The goodness-of-fit and model diagnostics of the model are also discussed. A cutaneous melanoma dataset example is used to illustrate the models as well as the inferential methods. Next, in Chapter 3, the destructive cure rate models, introduced by Rodrigues et al. (2011), are discussed under the PO assumption. Here, the initial number of competing causes is modelled by a weighted Poisson distribution with special focus on exponentially weighted Poisson, length-biased Poisson and negative binomial distributions. Then, a damage distribution is introduced for the number of initial causes which do not get destroyed. An EM-type algorithm for computing the MLEs is developed. An extensive simulation study is carried out for various scenarios, and model discrimination between the three weighted Poisson distributions is also examined. All the models and methods of estimation are evaluated through a simulation study. A cutaneous melanoma dataset example is used to illustrate the models as well as the inferential methods. In Chapter 4, frailty cure rate models are discussed under a gamma frailty wherein the initial number of competing causes is described by a Conway-Maxwell (COM) Poisson distribution in which the lifetimes of non-cured individuals can be described by PO model. The detailed steps of the EM algorithm are then developed for this model and an extensive simulation study is carried out to evaluate the performance of the proposed model and the estimation method. A cutaneous melanoma dataset as well as a simulated data are used for illustrative purposes. Finally, Chapter 5 outlines the work carried out in the thesis and also suggests some problems of further research interest. / Thesis / Doctor of Philosophy (PhD)

Indiana University-Purdue University Indianapolis (IUPUI) / Linking the new institutionalism to entrepreneurial orientation (EO), my dissertation investigates institutional forces and entrepreneurial forces—two contradicting types of forces—as main effects and moderating effects upon practices and performance of organizations embedded in the institutional duality. The case chosen observes unique hybrid funders that this study collectively calls social investment funders (SIF), which integrate philanthropy and venture capital investment to create and implement a venture philanthropy model for a pursuit of their mission. A theoretical framework is developed to propose regulative and normative pressures from two dominant institutions governing SIFs. Original data collected from 146 organizations are scrutinized by moderated multiple regressions for two empirical studies: Study 1 for effects on SIFs’ venture philanthropy practices, and Study 2 for effects on SIFs’ social and financial performance. Multiple imputations, diagnostic analyses, and several post hoc analyses are also conducted for robustness of data and results from multiple regression analyses. Results from these analyses find that EO and venture capital institutional forces both enhance SIFs’ venture philanthropy practices. A hypothesis postulated for a negative relationship between the nonprofit status and venture philanthropy practices is also supported. Results from moderated regression analyses, along with a subgroup and EO subdimension analyses, confirm a moderating effect between EO and the nonprofit status, i.e., a regulative institutional pressure. A positive relationship is found in EO- financial performance, but not in EO-social performance. While support is lent to hypotheses posited for a social/financial performance relationship with donors’/investors’ demand for social outcomes, and with the management team’s training in business, the overall results remain mixed for Study 2. Nonetheless, this dissertation appears to be the first study to theorize and test EO as a micro-level condition enabling organizations to strategically shape and resist institutional pressures, and it reinforces that organizations’ behavior is not merely a product of their passive conformity to environmental forces, but of the agency, also. As such, this study aims to contribute to scholarly efforts by the “agency camp” of the new institutionalism and EO, answering a call from the leading scholars of both EO (Miller) and the new institutionalism (Oliver).

En fiabilité et en analyse de survie, les distributions qui ont une fonction de hasard unimodale ne sont pas nombreuses, qu'on peut citer: Gaussienne inverse ,log-normale, log-logistique, de Birnbaum-Saunders, de Weibull exponentielle et de Weibullgénéralisée. Dans cette thèse, nous développons les tests modifiés du Chi-deux pour ces distributions tout en comparant la distribution Gaussienne inverse avec les autres. Ensuite nousconstruisons le modèle AFT basé sur la distribution Gaussienne inverse et les systèmes redondants basés sur les distributions de fonction de hasard unimodale. / In reliability and survival analysis, distributions that have a unimodalor $\cap-$shape hazard rate function are not too many, they include: the inverse Gaussian,log-normal, log-logistic, Birnbaum-Saunders, exponential Weibull and power generalized Weibulldistributions. In this thesis, we develop the modified Chi-squared tests for these distributions,and we give a comparative study between the inverse Gaussian distribution and the otherdistributions, then we realize simulations. We also construct the AFT model based on the inverseGaussian distribution and redundant systems based on distributions having a unimodal hazard ratefunction.