Orientador: Suzana Erico Tanni Minamoto / Resumo: Introdução: Os pacientes com doença pulmonar obstrutiva crônica (DPOC) exibem características clínicas heterogêneas que estão associados a diferentes respostas a tratamentos e prognósticos. A complexidade da doença faz com que sejam buscadas ferramentas alternativas como o agrupamento de cluster para a identificação de características específicas e que possam tratadas diferentemente dentro da mesma doença. Entretanto, ainda são escassos dados da América Latina em relação aos possíveis clusters da DPOC. Objetivo: Avaliar os possíveis clusters na DPOC em dois centros de estudo no Brasil. Métodos: Os pacientes foram submetidos à avaliação composta por doenças associadas, Índice de Charlson, composição corporal, fármacos atuais, história de tabagismo (anos/maço), monóxido de carbono exalado, histórico de exacerbações/ hospitalizações no último ano, espirometria, teste de caminhada de seis minutos, questionários de qualidade de vida, dispneia e escala hospitalar de ansiedade e depressão. Também foram coletadas amostras de sangue para dosagens de proteína C reativa (PCR), gases sanguíneos, análise laboratorial e hemograma. Resultados: Foram avaliados 334 pacientes portadores de sintomas respiratórios e fator de risco para DPOC. Desse total, 13 pacientes foram excluídos do estudo por não terminarem o protocolo, 20 pacientes sem diagnóstico de DPOC e 10 por não terem sido classificados em nenhum cluster. Assim, foram incluídos 291 pacientes [53,6% homem, 67,5 ± 9,6 anos e volume expi... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Introduction: Patients with chronic obstructive pulmonary disease (COPD) exhibit heterogeneous clinical features that are associated with different responses to treatments and prognoses. The complexity of the disease causes alternative tools such as clustering to identify specific characteristics and that can be treated differently within the same disease. However, there are still little data from Latin America regarding possible clusters in COPD. Objective: To evaluate possible clusters in COPD in two Brazilian centers. Methods: We assesses the comorbidities, Charlson's index, body composition, pharmacological treatment, smoking history (pack-years), exhaled carbon monoxide, exacerbations/hospitalizations rate in the last year, spirometry, six-minute walk test, quality of life questionnaires, dyspnea and hospital anxiety and depression scale. We also collected blood gases, laboratory and blood counts. Results: A total of 334 patients with respiratory symptoms and a risk factor for COPD were evaluated. From the total, 13 patients were excluded from the study because they didn’t complete the protocol, 20 patients without a diagnosis of COPD and 10 because they were not classified in any cluster. Thus, 291 patients were included [53.6% male, 67.5 ± 9.6 years and forced expiratory volume in the first second (FEV1) = 45.5 ± 17.9]. For the construction of the clusters, 13 continuous variables were selected and an analysis was performed with the Ward method and K method, which dete... (Complete abstract click electronic access below) / Mestre

A obtenção de previsões com maior acuracidade é uma necessidade constantemente requerida, em tempos onde há imensa disponibilidade de dados e recursos computacionais cada dia mais eficientes. Tais critérios possibilitaram o desenvolvimento de muitas técnicas de previsão individual ou de métodos de combinação que são considerados eficientes no intuito de reduzir erros. O desenvolvimento de novas técnicas, por sua vez, promove questionamentos quanto à identificação de quantas ou quais técnicas de previsão individual combinar. A literatura não é unânime ao tentar responder a estes questionamentos e indica a importância da correlação entre os erros de previsão na precisão da combinação. Posto isso, esta tese apresenta uma alternativa aos métodos atuais de combinar previsões, contemplando a correlação entre os erros de previsão, além de propor uma forma de identificar técnicas de previsão que sejam distintas quanto à modelagem de características da série de dados. Para identificar grupos de técnicas de previsão individual que sejam similares, utilizou-se a Análise de Agrupamentos em erros gerados por 15 técnicas de previsão que modelaram uma mesma série de dados real com tendência e sazonalidade. O resultado indicou a formação de 3 agrupamentos. Como alternativa aos métodos atuais de combinar previsão e selecionar a quantidade adequada de técnicas, utilizou-se a Análise de Componentes Principais. O método proposto mostrou-se viável quando comparado com outros métodos de combinação e quando submetido à modelagem de séries com maior variabilidade. / The obtaining of more accurate forecasts is a necessity often required in times where there is a huge availability of data and computing resources becoming more efficient every day. These criteria allowed the development of many individual forecasting techniques or combination methods that are considered efficient in order to reduce errors. The development of new techniques, in turn, promotes questioning as the identification of how many or which techniques to combine individual forecasts. The literature is not unanimous when trying to answer these questions and indicates the importance of the correlation between forecast errors on the accuracy of the combination. That said, this presents an alternative to current methods of combining forecasts, considering the correlation between forecast errors, and propose a way to identify predictive techniques that are different about the modeling features of the data series. To identify groups of individual forecasting techniques that are similar, it was used the cluster analysis on errors generated by 15 forecasting techniques that shaped the same series of real data with trend and seasonality. The result indicated the formation of 3 clusters. As an alternative to current methods of combining forecasting and selecting the appropriate amount of techniques, it was used the Principal Component Analysis. The proposed method has proved feasible when compared to other methods of combining and when subjected to modeling of series with greater variability.

The entrepreneur as an economic agent is highlighted as being important by several researchers, mainly through their contribution to innovation and economic development. Yet there is no universal definition of what is characterized as being an entrepreneur. The elements affecting success of entrepreneurs has previously not been outlined extensively, the aim of this study is therefore to determine the factors of entrepreneurial success in Sweden. This is made possible by adopting a statistical prosopography and reproducing the method used by Nuvolari, Toninelli & Vasta (2015). The method involves collecting qualitative information found in the Dictionary of Swedish National Biography and from a list of the most prominent entrepreneurs in Sweden, published by the newspaper Svenska Dagbladet. The qualitative information is then rewritten into quantitative data, which enables the use of factor analysis along with ordinary least square regressions. The standardized research allows for comparison of other studies involving entrepreneurs active in different countries. In addition, the model is expanded using a cluster analysis, allowing for identifying structures and patterns within the sample of entrepreneurs. The study concludes that entrepreneurial success appears to be a multi-dimensional phenomenon, also shown in other case studies adopting similar method. The level of innovation intensity proves to enhance success of all dimensions. While an entrepreneur’s educational level is shown to have a positive relationship with the economic success, it appeared to be negatively related to the aspect of social mobility. Moreover, involvement in politics is also shown to have a negative impact on economic success. Finally, the cluster analysis proved the existence of six different types of entrepreneurs within the sample.

O presente trabalho busca desenvolver um estudo comparativo entre três culturas distintas buscando identificar os valores pessoais dos consumidores de vinho orgânico. Percebe-se, atualmente, entre os países em desenvolvimento um grande avanço na internacionalização da comercialização de sua produção agroalimentar. Esses países estão cada vez mais conscientes de sua interdependência global e reconhecem neste contexto oportunidades de expansão de seus mercados. Estas oportunidades podem ser efetivas, uma vez que em países com alto grau de industrialização, as limitações para produção agrícola é uma realidade. Dentre as diferentes transformações que estão ocorrendo nos sistemas de produção e nos comportamentos de consumo, pode-se realçar a produção orgânica. Desta forma, este estudo tem como objetivo realizar um estudo comparativo entre consumidores de três diferentes nacionalidades (brasileira, francesa e espanhola), tendo como base seus valores pessoais. Tomou-se como base teórica para este estudo, conceitos relacionados principalmente à internacionalização (BASSI, 1997), cadeias produtivas (CHRISTOPHER, 1997), cultura (HOFSTEDE, 1980) e valores pessoais (KAHLE, 1983). Com isso, identificou-se que o método mais adequado a ser utilizado nesta pesquisa seria desenvolvido em duas etapas. A primeira etapa teve como objetivo fornecer subsídios para a elaboração do questionário aplicado. Foram realizadas vinte e cinco entrevistas nas localidades pesquisadas no Brasil (Porto Alegre), na França (Montpellier) e na Espanha (San Sebastián), com especialistas sobre produção e mercado de produtos orgânicos, com consumidores de vinho orgânico bem como com intermediários que comercializam o vinho orgânico e puderam contribuir com a experiência no contato com os produtores e consumidores. A segunda etapa da pesquisa se deu por meio da coleta dos dados a partir de questionário elaborado em função dos resultados da primeira etapa. O questionário foi desenvolvido em português e posteriormente traduzido para o francês e espanhol por profissionais especializados. A análise dos dados se deu através de Análise de Cluster, o que permitiu reconhecer os comportamentos das culturas distintas, além de identificar questões em comum. Entre os 3 clusters identificados, observou-se que o Cluster 1, classificado como Coletivistas, apresenta características de mercado de massa, não percebendo valor suficiente no vinho orgânico para justificar o pagamento de um preço mais elevado. Apesar de serem orientados para questões coletivas, não percebem a proposta de amenizar o impacto no ambiente e a valorização do produtor rural como argumentos para aquisição. Os Individualistas, cluster 2, apresentaram uma preocupação muito grande com as suas realizações pessoais. Mostraram envolvimento moderado com as questões ambientais e procuram desfrutar dos seus recursos em benefício próprio. Há uma grande proximidade entre este cluster e as características da cultura francesa. O Cluster 3, classificado como Afetuosos, devido ao valor pessoal considerado mais importante, se mostra mais acessível aos argumentos da produção orgânica, apresentando orientações bastante individualistas, mas valorizando principalmente o Relacionamento caloroso com os outros. Os seus integrantes reconhecem valor no vinho orgânico e aceitam pagar um preço mais elevado por isso. Para uma proposta de manter o vinho orgânico num posicionamento de status com diferencial, este cluster é o mais adequado para ser trabalhado. Com estes resultados, foi possível desenvolver um conjunto de propostas de atuação por parte da cadeia produtiva do vinho orgânico visando os mercados local e internacional. / This work aims to develop a comparative study among three distinct cultures trying to identify the personal values of the consumers of organic wine. Currently, among developing countries, it is possible to note a marked advance in the internationalization of the commercialization of their food production. These countries are ever more aware of their global interdependence and the opportunities for the expansion of their markets. Such opportunities may become real, given that, in highly industrialized countries there are limitations of agricultural production. Among the different transformations taking place in production systems and consumer behavior, organic production is of particular note. This way, this study aims to do a comparative study among consumers of three different nationalities (Brazilian, French and Spanish) using as a base their personal values. The study consisted of two stages. The aim of the first stage was to supply subsidies for the elaboration of the applied questionnaire. In Brazil (Porto Alegre), France (Montpellier) and Spain (San Sebastián), twenty-five interviews were carried out with specialists in the production of, and market for organic products and consumers of organic wine, as well as with traders of organic wine, who were able to contribute with their experience of contact with both producers and consumers. The second stage of the study consisted of the collection of data obtained from a questionnaire elaborated based on the results from the first stage. The questionnaire was prepared in Portuguese and then translated into French and Spanish by specialized professionals. The analysis of the data was made through Cluster Analysis with allowed to recognize the behaviors of the distinct cultures, as well as identifying questions in common. Among the 3 clusters identified, it was observed that Cluster 1, classified as Collectives, has some characteristics of mass market and its components don’t perceive value in the organic wine to justify the payment of a raised price. Although to be guided for collective questions they don’t understand the proposal to reduce the impact in the environment and the valuation of the agricultural producer as arguments for acquisition. The Individualists, cluster 2, are concerned with its personal accomplishments. They show a moderate involvement with the ambient issues and try to enjoy its resources in proper benefit. It is a great proximity of this cluster to the characteristics of the French culture. Cluster 3 is classified as Affectionate, due to the indication of Warm relationship with the others as the most important personal value. It seems to be more accessible to the arguments of the organic production, spite of their individualistic orientations. Its integrants perceive value in the organic wine and accept to pay a higher price for this. Considering the aim of to keep the organic wine in a positioning of status with differential, this cluster is indicated. The results of this study permit the development of a set of proposals of performance on the organic wine chain productive aiming at the local and international markets.

The study is concerned with effective tax rates, namely implicit tax rates in the European Union over the period 2000-2012. The theoretical part includes tax incidence, description of individual implicit tax rates and their structure, and Eurostat metodology for measuring the impact of taxation on economic functions. The practical part in mainly dedicated to the development of implicit tax rates in the European Union and then to the evaluation of development trends in the implicit rates on consumption, labour and capital. The analysis is processed in the computer program STATISTICA. The classify is made by cluster analysis that should find the similar implicit tax rates in 2000, 2008 and 2012. Subsequently, the trends ort he development of individual implicit tax rates in EU countries are evaluated.

The objective of the thesis is to investigate the development of excise taxes indicators in twenty-seven member states of the European Union (without Croatia due to the absence of data). The selected period for the analysis is the year 2004 - 2016. The main aim of the thesis is to define the groups of member states EU according to their importance and similarity with respect to the excise duties. With this aim is related the another important idea of the work. That is whether the indicators of excise duties are brought closer to the harmonization processes and if are the member states converging. The cluster analysis was used to achieve the main goal of the work. The data which is used in this analysis is obtained from Eurostat and European Commission databases. The results of cluster analyzes allow the member states to be divided into four of the most stable groups. The group of Nordic countries in Europe, group of border states, states of NMS-12 able to approach the EU-15 and other states of NMS-12. The cluster analyzes also shows that the original and new member states, that joined the EU in 2004 and later, are still different in the field of excise duties. Differences that exist despite the strict harmonization of individual tax systems, in particular the harmonization of excise rates, may be caused, in particular, to the historical roots of individual countries, but also by the economic and political situation.

The purpose of this study is to replicate and extend earlier work involving cluster analysis of MMPI profiles among persons with chronic low back pain. There are two specific goals. The first goal is to demonstrate the existence in a new sample of four distinct and homogenous profile clusters that have been found in previous research. The second goal is to investigate the relationship of the four profiles to the subjects, self-reported pain history and response to treatment. This study concludes that four distinct MMPI profiles can be identified among chronic low back pain patients. Further, these profiles are the same for males and females, and are the same profiles found in previous research. These profiles are significantly related to subjects' history of behaviors in dealing with pain. However, no relationship to treatment response was found. It was inferred that the MMPI is of value in understanding the nature of patients' pain coping behaviors, but that further research is needed before any statements can be made regarding the utility of the MMPI in understanding their response to treatment.

Made available in DSpace on 2015-03-26T13:32:18Z (GMT). No. of bitstreams: 1 texto completo.pdf: 765144 bytes, checksum: 57b7456f9464c0ead5c7fb45295d0b99 (MD5) Previous issue date: 2013-02-06 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Cluster analysis is widely used in many research areas in order to recognize a standard structure of variability between individuals or objects studied, classifying them into homogeneous groups. However, the studies that are published, most of them deal only on numeric variables, excluding the analysis, the information contained in categorical variables. Thus, this study aims to evaluate some similarity measures and clustering algorithms in databases and also simulated on a case study in Genetics. The similarity measures evaluated were: euclidean, squared euclidean, mean euclidean, mahalanobis, manhattan, combined measures and gower. The hierarchical clustering algorithms are: nearest neighbor, furthest neighbor, UPGMA and Ward. The algorithms evaluated from the class of non-hierarchical are the kmeans and k-prototypes, which is an extension of the first. The results were compared and we concluded the non-hierarquical were better than hierarquical methods. / A análise de agrupamento é amplamente utilizada em muitas áreas de pesquisa a fim de se reconhecer uma estrutura padrão de variabilidade entre os indivíduos ou objetos estudados, classificando-os em grupos homogêneos. No entanto, dos trabalhos publicados, a maioria deles versam apenas sobre variáveis numéricas, excluindo da análise, as informações contidas nas variáveis categóricas. Dessa forma, esse trabalho teve o objetivo de avaliar várias formas de agrupamentos em um banco de dados simulado e também de disponibilizar uma rotina em R do algoritmo kprotótipos e uma rotina para se realizar agrupamentos hierárquicos. As medidas de distâncias avaliadas foram: euclidiana, euclidiana ao quadrado, euclidiana média, mahalanobis, manhattan, medidas combinadas e a de gower. Quanto aos algoritmos de agrupamento hierárquicos utilizados foram: vizinho mais próximo, vizinho mais distante, UPGMA e ward . Os algoritmos não-hierárquicos foram: k-médias e o kprotótipos. Os resultados obtidos foram confrontados entre si e concluiu-se que os algoritmos não-hierárquicos foram superiores aos hierárquicos e que incluir variáveis categóricas na análise é viável.

Social networks tend to shape our views about the world. Our study conducts an empirical analysis of social network dynamics using Twitter data. We ask whether social networks influence voting decisions, and determine whether or not people make consistent choices based on their tweets or what they believe. We collect Twitter data on a daily basis, with dynamic social network measurements before, during, and after the 2012 Presidential election. We identify how people should believe based on their ideological profiles. We use lexicographical analysis to check if ideological key words are present in a user's tweets, and if the overall sentiment on this issue is positive or negative. We utilize this data to determine how people should have chosen an outcome which may conflict with an individual's observed declaration of political ideology. We are able to determine what percentage of the population made a consistent choices based on their Tweets during the 2012 presidential election. Additionally, we examine the social network structure in Twitter and how it affects voting. We illustrate that an individual's political ideology is influenced by others in their network.\\ Consumer confidence is an economic indicator which measures the degree of optimism that consumers feel about the overall state of the economy as well as their personal financial situation. We will show that consumer sentiment can be measured via analysis of social networks. Specifically, we perform a lexicographic analysis of Twitter data over a three month period. After careful analysis, we find that not only does talk intensity of economic issues cause shifts in the daily stock market prices but has a significant negative affect.\\ The study of religion has enjoyed distinction and legitimacy within sociology, psychology, anthropology, and political science for many years. This paper concerns the extent to which economic opinion is embedded in structure of religious social relations. We hope to enhance the empirical study of homophily and the economics of behavior by showing how beliefs, norms, and values are affected by religion and, by extension, morals, and culture. We utilize a technique called cluster analysis to determine homophilic ties within a single attribute, religiosity. We see that religion affects economic attitudes and activities of individuals, groups, and societies. Further, religion influences how behavior and institutions are affected by social relations and in our case homophily. This influence is one of the classic questions of social theory.

Traditional forecasting approaches forecast the total system load directly without considering the individual consumer's load. With the introduction of the smart grid, lots of renewable energy resources such as wind and solar are added to the system from consumer side fluctuates the system load and makes forecasting more complex. Thus, it is necessary to forecast individual consumers load. Here, a framework is presented in which individual customer loads is forecasted rather than the system load. At first, a hierarchical cluster analysis is performed to classify daily load patterns into different groups for all the individuals. Then an association analysis is performed to determine critical influential factors that affect the load curve for given day. The next step is the application of a decision tree to establish classification rules between the different groups of the load curve and the critical influential factors. Then, appropriate forecasting models are chosen for different load patterns and the individual load is forecasted. Finally, the forecasted total system load is obtained through an aggregation of an individual load forecasting results. The relative error of forecasting the system load using this framework is compared with the relative errors using SVM regression and this framework had better accuracy. This framework is also used for forecasting the power output of the renewable generation. Also, the results of the day ahead forecast of system load and renewable generation is used for economic power scheduling for the microgrid and peak shaving for the utilities.