Análise de dados e aplicação de modelos estatísticos para apoio à decisão de contratação na Petróleo Brasileiro

Dissertação de mestrado em Engenharia Industrial

Dados estatísticos de viscerotomia sobre doenças e condições mórbidas do homem no Brasil: I. Schistosomose Mansônica no período de 1937-1946

Examination of 267.107 liver specimens obtained in Brazil by viscerotomy during from 1937 to 1946 inclusive revealed 5,953 Schistosoma mansoni infe¬ctions. This represents 2.23% ± 0.019 of the total number of livers studied. Data on the incidence of the disease is tabulated by states and municipios. Infected livers were found in all of the states and territories except the Territory of Amapá. Schistosomiasis is widespread in Brazil with highest incidence in the states of the Northeast. The disease is quite common in Espírito Santo and Minas Gerais as well. A study of the age distribution of cases of intestinal schistosomiasis observed among liver specimens obtained in the year 1938 showed a low inci¬dence on young children with a peak of prevalence in the 10 to 19 year age group. The purpose of this contribution is to call attention of the health autho¬rities to the extent and gravity of the problem of intestinal schistosomiasis in Brazil.

Indicadores Estatísticos na Administração Escolar Estudo de Caso: Escolas Secundárias/ Liceus da ilha de Santiago

Imbuídos da melhor das intenções, idealizamos um trabalho que contribuísse, ainda que modestamente, para melhorar a relação que alguns dos nossos colegas do curso de Ciências da Educação e Praxis Educativa, vertente Administração Escolar, tinham com a Estatística A tradução dessa ideia em algo concreto é este documento que agora apresentamos. Nele começamos por apresentar uma definição de Estatística, para logo a seguir apresentarmos algumas ferramentas estatísticas muito práticas, fáceis de entender e de utilização imediata na esfera de actuação do Administrador escolar e com exemplos de como manuseá-los. Depois apresentamos alguns indicadores que, em nosso entender, devem fazer parte não só do vocabulário de qualquer administrador escolar, como também do leque de factores que o orientam e nos quais deverão alicerçar-se as decisões que tiver que tomar. Finalmente fizemos uma análise dos dados recolhidos, seja através de questionários, seja através de entrevistas, junto dos directores das escolas secundárias, dados que espelham o relacionamento dos actuais detentores do cargo com a Estatística.

UM OLHAR SOBRE OS ANUÁRIOS ESTATÍSTICOS

Este texto lança um olhar sobre os anuários estatísticos. Após uma breve nota histórica, prossegue com a declaração de seus objetivos gerais e específicos. Depois trata de sua estrutura, primeiro sob uma visão geral, depois sob uma visão específica. Por fim, trata do público e da disseminação de um anuário. Finalmente, aborda a delicada questão de como criticar um anuário com vistas à sua reformulação permanente

Prospective sampling based on model ensembles improves the detection of rare species

Identifying the geographic distribution of populations is a basic, yet crucial step in many fundamental and applied ecological projects, as it provides key information on which many subsequent analyses depend. However, this task is often costly and time consuming, especially where rare species are concerned and where most sampling designs generally prove inefficient. At the same time, rare species are those for which distribution data are most needed for their conservation to be effective. To enhance fieldwork sampling, model-based sampling (MBS) uses predictions from species distribution models: when looking for the species in areas of high habitat suitability, chances should be higher to find them. We thoroughly tested the efficiency of MBS by conducting an important survey in the Swiss Alps, assessing the detection rate of three rare and five common plant species. For each species, habitat suitability maps were produced following an ensemble modeling framework combining two spatial resolutions and two modeling techniques. We tested the efficiency of MBS and the accuracy of our models by sampling 240 sites in the field (30 sitesx8 species). Across all species, the MBS approach proved to be effective. In particular, the MBS design strictly led to the discovery of six sites of presence of one rare plant, increasing chances to find this species from 0 to 50%. For common species, MBS doubled the new population discovery rates as compared to random sampling. Habitat suitability maps coming from the combination of four individual modeling methods predicted well the species' distribution and more accurately than the individual models. As a conclusion, using MBS for fieldwork could efficiently help in increasing our knowledge of rare species distribution. More generally, we recommend using habitat suitability models to support conservation plans.

Métodos estatísticos para estudo de adaptabilidade e estabilidade fenotípica em soja

O objetivo deste trabalho foi avaliar os métodos estatísticos de análise da interação de genótipos com ambientes (GxA), enfatizando a adaptabilidade e a estabilidade fenotípica. Utilizaram-se dados de produtividade de grãos de soja de sete experimentos em Goiás, testando 28 genótipos, dos quais quatro cultivares comerciais. Avaliaram-se os métodos Tradicional, Plaisted & Peterson, Wricke, Finlay & Wilkinson, Eberhart & Russell, Verma, Chahal & Murty, Toler, AMMI (additive main effect and multiplicative interaction), Hühn, Annicchiarico e Lin & Binns. Avaliou-se a associação entre os métodos pela correlação de Spearman. Observou-se forte associação entre os de Plaisted & Peterson e Wricke, cujo uso concomitante foi contra-indicado. A mesma conclusão é atribuída aos métodos Annicchiarico e Lin & Binns, também fortemente associados, o que implica em classificações fenotípicas muito semelhantes. O uso de um deles, entretanto, é recomendado. Métodos baseados, exclusivamente, em coeficientes de regressão, devem ser utilizados em associação com outro, fundamentado na variância da interação GxA, ou em medidas estatísticas como a variância dos desvios da regressão. O uso combinado do método de Eberhart & Russell e AMMI é outra indicação, em razão de suas correlações significativas com a maioria dos outros métodos e uma associação relativamente fraca entre eles.

Réflexions sur les céramiques fines en Gaule interne de La Tène finale au début du Haut-Empire : comparaisons d'ensembles et essai de caractérisation des phénomènes d'acculturation

Le concept de romanisation ne revêt d'utilité que s'il peut être qualifié, et donc analysé en différents « niveaux », afin de comparer les rythmes et les formes du phénomène dans les régions de l'Empire. Avant même de chercher à tenir un discours global sur la romanisation, il faut donc en étudier les multiples aspects de manière spécifique. C'est dans ce cadre que s'inscrit cette thèse de doctorat dont le but est de quantifier et de qualifier l'acculturation des céramiques fines en Gaule interne (200 BC - 50 AD). Fondée sur un corpus de quelques 130 ensembles, provenant de 21 sites urbains localisés entre le centre ouest de la France, l'Auvergne, le Plateau suisse et le Luxembourg, cette étude se caractérise par la mise en place de nouvelles approches méthodologiques, s'appuyant sur le calcul d'un « indice de romanisation » (quantifiant la romanisation d'un ensemble sur une échelle de 1 à 100) et sur l'analyse des modalités du phénomène. L'examen diachronique et cartographique des résultats obtenus montrant de fortes disparités dans les processus d'acculturation des céramiques fines, des niveaux de romanisation ont pu être définis, dont le principal intérêt est d'ouvrir la comparaison avec les autres aspects de l'acculturation des Gaules (instrumentum culinaire, technique de construction, romanisation de l'anthroponymie...). The concept of Romanisation can only be considered useful if it can be qualified and analysed at different levels in order to compare the acculturation rates, frequency and form of the phenomenon among the regions in the Empire. But before an overview of Romanisation can be considered, its multiple aspects must be examined in a specific manner. Among the French speaking areas, it is only recently that research concerning Gaulish and Roman periods has risked expanding the scope to a supra regional scale. Against this background, the purpose of my doctoral thesis is to examine the qualitative and quantitative aspects of acculturation among Gaulish fine wares (200 BCE - 50 CE). Based on a corpus of 130 assemblages from 21 urban sites from the western centre of France, Auvergne, the Swiss plateau and Luxembourg, the study is characterised by new methodological techniques relying on the calculation of a 'Romanisation index' (quantifying the level of Romanisation of an assemblage on a scale of 1 to 100) and the analysis of the methods of the phenomenon. The diachronic and cartographic examination of the results obtained showed large disparities in the acculturation processes concerning fine wares. The level of Romanisation could be defined, thus allowing a comparison with other aspects of the acculturation process among other aspects of Gaulish culture such as tools, cuisine, construction techniques and Romanisation of anthroponyms.

Overcoming limitations of modelling rare species by using ensembles of small models

1. Species distribution models (SDMs) have become a standard tool in ecology and applied conservation biology. Modelling rare and threatened species is particularly important for conservation purposes. However, modelling rare species is difficult because the combination of few occurrences and many predictor variables easily leads to model overfitting. A new strategy using ensembles of small models was recently developed in an attempt to overcome this limitation of rare species modelling and has been tested successfully for only a single species so far. Here, we aim to test the approach more comprehensively on a large number of species including a transferability assessment. 2. For each species numerous small (here bivariate) models were calibrated, evaluated and averaged to an ensemble weighted by AUC scores. These 'ensembles of small models' (ESMs) were compared to standard Species Distribution Models (SDMs) using three commonly used modelling techniques (GLM, GBM, Maxent) and their ensemble prediction. We tested 107 rare and under-sampled plant species of conservation concern in Switzerland. 3. We show that ESMs performed significantly better than standard SDMs. The rarer the species, the more pronounced the effects were. ESMs were also superior to standard SDMs and their ensemble when they were independently evaluated using a transferability assessment. 4. By averaging simple small models to an ensemble, ESMs avoid overfitting without losing explanatory power through reducing the number of predictor variables. They further improve the reliability of species distribution models, especially for rare species, and thus help to overcome limitations of modelling rare species.