Submitted by Luciana Ferreira (lucgeral@gmail.com) on 2015-04-23T14:13:46Z No. of bitstreams: 2 Dissertação - Tiago Carvalhais de Oliveira - 2014.pdf: 2283233 bytes, checksum: 6e0744d8d8b25db17ecedc5ed707eee8 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2015-04-23T14:16:09Z (GMT) No. of bitstreams: 2 Dissertação - Tiago Carvalhais de Oliveira - 2014.pdf: 2283233 bytes, checksum: 6e0744d8d8b25db17ecedc5ed707eee8 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2015-04-23T14:16:09Z (GMT). No. of bitstreams: 2 Dissertação - Tiago Carvalhais de Oliveira - 2014.pdf: 2283233 bytes, checksum: 6e0744d8d8b25db17ecedc5ed707eee8 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2014-02-28 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Caterpillars defoliators of soybean occurs in different times and feed on different parts of the plant, mainly in flowers and pods. During the growing seasons of 2011/2012 and 2012/2013, soybean fields were sampled in different crop phenological stages with the objective of knowing the population fluctuation of the main species of caterpillar defoliating soybeans and to determine the within field spatial distribution of the species of the Plusiinae complex, to describe how species colonize the soybean field. Five commercial fields submitted to different management practices were monitored. Sampling meshes were fabricated with equidistant points 50 meters. The areas were sampled weekly from the occurrence of the first defoliating caterpillars, until close to harvest. The sampling method used was ground cloth, with one row-meter. It was quantified all caterpillars. Data were submitted to statistical analysis using the following SAS procedures: Proc Univariate, Proc Means, Proc Freq. To test what statistical distribution fit to caterpillars population the procedure Proc Genmode was used. Maps representing the localization of samples of caterpillars in soybean fields were constructed. The species of Plusiinae predominated in the area, corresponding to 52, 04% in 2011/2012 e 48, 53% in 2012/2013 of the total of caterpillars found in the reproductive phase of soybean. The distribution of the complex of Plusiinae within field is random and do not present border effect. In this study, while the population fits the negative binomial distribution maps clearly indicate that the caterpillars are randomly dispersed in soybean fields. / As lagartas desfolhadoras da soja ocorrem em épocas distintas e atacam várias partes da planta, principalmente as folhas, flores e frutos. Durante os anos agrícolas de 2011/2012 e 2012/2013, campos de soja foram amostrados em diversos estádios fenológicos da planta com o objetivo de conhecer a flutuação populacional das principais espécies de lagartas desfolhadoras da cultura, e determinar a distribuição espacial em lavoura, espécies pertencentes ao complexo de Plusiinae, para compreender a forma que este complexo coloniza os campos de soja. Foram monitoradas cinco lavouras comerciais, submetidas a diferentes condições de manejo. Foram confeccionadas malhas de amostragem, com pontos eqüidistantes de 50 metros. As áreas foram amostradas semanalmente a partir da ocorrência das primeiras lagartas desfolhadoras, até próximo à colheita. O método de amostragem utilizado foi o pano de batida, com o qual um metro linear foi amostrado. Foram quantificadas todas as lagartas presentes nas amostragens dos campos selecionados. Para a realização das inferências, dos dados foram submetidos á análises estatísticas descritivas utilizando os seguintes procedimentos dos Sas, Proc Univariate, Proc Means e Proc Freq. Para verificar qual a distribuição estatística a que a população de lagartas melhor se ajustava foi utilizado o procedimento Proc Genmode. Mapas representativos da localização das lagartas do complexo Plusiinae no campo foram confeccionados. As espécies de Plusiinae foram as lagartas predominantes, compondo 52,04% em 2011/12 e 48,53% em 2012/13, do total de lagartas encontradas. As flutuações populacionais demonstraram que o pico das lagartas desfolhadoras ocorre principalmente na fase reprodutiva da cultura da soja. A distribuição de lagartas do complexo Plusiinae em lavoura é aleatória, não apresentando efeitos de bordas. Neste estudo, enquanto a população se ajusta à distribuição binomial negativa os mapas indicam claramente que as lagartas estão aleatoriamente dispersas nos campos de soja.

Abstract Mortuary rituals express and cope with disorder brought about by a member's death in the community. The autonomous connection of the deceased with the community is disrupted through mortuary rituals. In many cultures the subsequent contacts with the realm of the dead are maintained in formalized practices, sometimes including or referring to objects or patterns that can be traced in the archaeological record. In this study it is asked, if the Bronze Age and Iron Age burial cairns (1200 BC - AD 1000) in the SW archipelago of Finland might be interpreted as monuments establishing a link between the landscape and the religious context of symbolic meanings, thus making it meaningful to examine the spatial references of grave sites. The field studies include excavations, surveys, boulder analyses, and weathering studies. The number of cairns in the area is 444. Examination of samples of boulders suggested that the stones were usually collected from the adjacent terrain. The Schmidt hammer technique was applied to measure the weathering differences between basal and lateral surfaces, and possible secondary interference. The chronology of the archipelago cairns is based on previous studies related to general chronological characteristics and datings of archipelago graves. Using discriminant analysis, the size of the cairn, the convexity of the surface at the grave site, and the topography of the terrain were identified as the variables most related to the differences between Group P, having a Bronze Age character (147 cairns), from Group R of Iron Age character (218 cairns). Two models representing the shorelines of 500 BC and AD 1000 were reconstructed using a digital elevation model (DEM). Monte Carlo-testing was applied when the visible areas around grave sites were compared to reference sets in four subareas. The grave sites in Group P were often directed towards the land, whereas the grave sites in Group R were typically directed towards the sea. The difference might be related to differences in subsistence strategies. The cairns represented a conservative burial custom that belonged to local communities in maritime and northern areas, as opposed to the southern agricultural environments. / Tiivistelmä Vainajan omaehtoinen yhteys elävien yhteisöön katkeaa vasta yhteisöllisen kuolemanrituaalin lopullisesti päätyttyä. Monissa kulttuureissa kuolemanrituaalin jälkeiset yhteydet vainajaan kiteytyvät muodollisiksi käytännöiksi, jotka voivat tulla arkeologisesti näkyviin aineellisissa jäännöksissä tai luonnonmaiseman paikkojen, tilojen ja elementtien suhteissa. Työssä tarkastellaan, ovatko Turunmaan saariston pronssikauden ja rautakauden hautarauniot (1200 e.Kr. - 1000 j.Kr.) tulkittavissa monumenteiksi, jotka yhdistivät maiseman symbolisten merkitysten uskomukselliseen kontekstiin. Kenttätutkimuksiin kuuluu kaivauksia, inventointi, lohkaretutkimuksia ja rapautumismittauksia. Hautoja on 444. Lohkaretutkimukset osoittivat kivien tulleen kerätyiksi hautapaikkojen läheisyydestä. Tapaustutkielmissa kiveyksen basaali- ja lateraalipintojen välistä rapautumiseroa ja sekundaarisia vaurioita tutkittiin kimmovasaramittauksin. Hautaraunioiden kronologia perustuu aikaisempiin tutkimuksiin kronologisista tunnusmerkeistä sekä saariston ajoitettuihin hautoihin. Erotteluanalyysissa kiveyksen laajuus, hautapaikan maanpinnan kuperuus ja hautapaikan suhde ympäröiviin huippuihin osoittautuivat muuttujiksi, jotka selvimmin jakavat aineiston pronssikauden tyypin P-ryhmään (147 hautaa) ja rautakauden tyypin R-ryhmään (218 hautaa). Numeerisesta korkeusmallista laskettiin kaksi maastomallia, jotka vastaavat rannansiirtymisen kehitysvaihetta 500 e.Kr. (P-ryhmä) ja 1000 j.Kr. (R-ryhmä). Hautapaikoilta näkyvissä olleita alueita verrattiin satunnaisesti valittuihin verrokkipaikkoihin Monte Carlo -testauksen avulla. Merkittävin ero oli, että P-ryhmän hautapaikat olivat tyypillisesti suuntautuneet merta ja R-ryhmän hautapaikat maata kohti. Ero liittynee toimeentuloon latautuneisiin odotuksiin ja epävarmuuksiin. Hautarauniot merkitsevät konservatiivista hautaustapaa, joka kuului enemmän mereisten ja pohjoisten paikallisyhteisöjen kuin agraarisen ja eteläisen asutuksen piiriin.

在許多實際情形下，傳統的統計檢定方法是不足以應付的。故本論文提出模糊檢定方法，我們定義出模糊樣本期望值與模糊樣本變異數的計算方法，再針對不同的模糊資料，分別提出不同的檢定方法，去解決最實際需要解決的問題，其中包括推廣古典的統計檢定方法與自創的檢定方法。 關鍵字：隸屬度函數，模糊樣本取樣，模糊樣本期望值，模糊樣本變異數，人性思考，t檢定，F檢定，模糊常態分配。 / In many expositions of fuzzy methods, fuzzy techniques are described as an alternative to a more traditional statistical approach. In this paper, we present a class of fuzzy statistical decision process in which testing hypothesis can be naturally reformulated in terms of interval-valued statistics. We provide the definitions of fuzzy mean, fuzzy distance as well as investigation of their related properties. We also give some empirical examples to illustrate the techniques and to analyze fuzzy data. Empirical studies show that fuzzy hypothesis testing with soft computing for interval data are more realistic and reasonable in the social science research. Finally certain comments are suggested for the further studies. We hope that this reformation will make the corresponding fuzzy techniques more acceptable to researchers whose only experience is in using traditional statistical methods. Key words: Membership function, fuzzy sampling survey, fuzzy mean, human thought, t-test, F-test, normally distributed.

在資料分析上，調查者通常需要決定，不同的樣本是否可被視為來自相同的母體。一般最常使用的統計量為Pearson’s 統計量。然而，傳統的統計方法皆是利用二元邏輯觀念來呈現。如果我們想要用模糊邏輯的概念來做樣本調查，此時，使用傳統 檢定來分析這些模糊樣本資料是否仍然適當？透過這樣的觀念，我們使用傳統統計方法，找出一個能處理這些模糊樣本資料的公式，稱之為模糊 。結果顯示，此公式可用來檢定，模糊樣本資料在不同母體下機率的一致性。 / In the analysis of research data, the investigator often needs to decide whether several independent samples may be regarded as having come from the same population. The most commonly used statistic is Pearson’s statistic. However, traditional statistics reflect the result from a two-valued logic concept. If we want to survey sampling with fuzzy logic concept, is it still appropriate to use the traditional -test for analysing those fuzzy sample data? Through this concept, we try to use a traditional statistic method to find out a formula, called fuzzy , that enables us to deal with those fuzzy sample data. The result shows that we can use the formula to test hypotheses about probabilities of various outcomes in fuzzy sample data.

Made available in DSpace on 2011-05-04T12:42:06Z (GMT). No. of bitstreams: 0 Previous issue date: 2009 / Dados nacionais sobre perfil de doenças e utilização de serviços odontológicos são raros no Brasil. O levantamento epidemiológico SB Brasil (Saúde bucal da população brasileira), realizada entre 2002 e 2003 foi a mais abrangente pesquisa sobre as condições de saúde bucal realizada até hoje, contemplando tanto aspectos relativos às doenças bucais quanto ao uso dos serviços odontológicos. O objetivo desta tese de doutorado foi identificar fatores associados às desigualdades na utilização dos serviços odontológicos no Brasil nas faixas etárias entre 15 a 19,35 a 44 e 65 a 74 anos, à partir do banco de dados do SB Brasil 2003. No entanto, para a realização deste objetivo, foi necessário incorporar ao banco de dados deste inquérito, as informações estruturais do desenho de sua amostra, uma vez que o processo de amostragem probabilística não foi concluído. Esta tese está dividida em três artigos: O primeiro que identifica e detalha os problemas encontrados no banco de dados do inquérito SB Brasil 2003, destacando a importância do cálculo dos pesos amostrais como fundamentais para a produção de dados não enviesados em uma amostra complexa e propõe formas de correção. O segundo que descreve as atividades de recuperação das informações necessárias ao cálculo dos pesos amostrais, o método utilizado para o cálculo dos referidos pesos e a técnica de calibração dos pesos. E, um terceiro artigo que investiga os fatores associados às desigualdades na utilização dos serviços odontológicos nasfaixas etárias de 15 a 19, 35 a 44 e 65 a 74 anos, por meio de análise multivariada. Os pesos amostrais e as demais variáveis estruturais do desenho da amostra da pesquisa SB Brasil 2003 foram determinados considerando as informações disponíveis sobre a seleçãodas unidades nos municípios incluídos, e uma nova versão do banco de dados foi produzida. Análises sobre o uso dos serviços odontológicos no Brasil foram então conduzidas, com modelos explicativos da sua variação apontando desigualdades entre adultos e idosos,sobretudo. Adicionalmente, diferenças encontradas nas análises realizadas com o banco de dados original e o banco corrigido ratificam a importância do uso dos pesos amostrais para a obtenção de estimativas confiáveis, provendo informação com maior qualidade dos dados nacionais sobre a saúde bucal da população brasileira. / National data about oral health problems and use of oral health services are rare in Brazil. The epidemiological survey SB Brazil (Oral Health Conditions of the brasilian population) was held between 2002 and 2003. This survey was the most comprehensive research on oral health conditions conduced until today, contemplating aspects about oral illnesses and dental services. The aim of this thesis was to identify factors associates to the inequalities in the use of the oral health services in Brazil in three age’s groups (adolescents, adults and elderly) from the data base of research SB Brazil 2003. However, for the accomplishment of this objective, it was necessary to incorporate the data base of this research, the structural information of the drawing of its sample, a time that the process of probabilistic sampling was not concluded. This thesis is divided in three articles: The first one that it identifies and it details the problems found in the data base of SB Brazil 2003 research, detaching the importance of the calculation of the weights sampling as basic for the production of data without bias in a complex sample and to consider correction forms. The second one describes the activities of recovery of the necessary information to calculate the sampling weights, the method used for the calculation of the weights and the technique of calibration of the weights. And, one third article that investigates the factors associates to the inaqualities in the use of the dental services in the age’s groups (15-19, 35-44 and 65-74 years old) by means of a multivaried analysis. The sample weights and others structural variables of the drawing of its sample were determined considering the available information on the selection of units in the municipalities included, and a new version of data base was produced. After, analysis on the use of dental services in Brazil were then conducted, with explanatory models of its variation pointing inequalities between adults and the elderly, especially. In addition, differences found in tests carried out with the original database and the bank fixed confirm the importance of using the sample weights to obtain reliable estimates, providing information with higher quality of national data on oral health of the population.