• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 111
  • 40
  • 21
  • 10
  • 8
  • 6
  • 4
  • 3
  • 2
  • 2
  • 2
  • 2
  • 1
  • Tagged with
  • 253
  • 63
  • 54
  • 34
  • 34
  • 30
  • 27
  • 23
  • 22
  • 22
  • 22
  • 21
  • 20
  • 20
  • 19
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
171

$L_\infty$-Norm Computation for Descriptor Systems

Voigt, Matthias 15 July 2010 (has links) (PDF)
In many applications from industry and technology computer simulations are performed using models which can be formulated by systems of differential equations. Often the equations underlie additional algebraic constraints. In this context we speak of descriptor systems. Very important characteristic values of such systems are the $L_\infty$-norms of the corresponding transfer functions. The main goal of this thesis is to extend a numerical method for the computation of the $L_\infty$-norm for standard state space systems to descriptor systems. For this purpose we develop a numerical method to check whether the transfer function of a given descriptor system is proper or improper and additionally use this method to reduce the order of the system to decrease the costs of the $L_\infty$-norm computation. When computing the $L_\infty$-norm it is necessary to compute the eigenvalues of certain skew-Hamiltonian/Hamiltonian matrix pencils composed by the system matrices. We show how we extend these matrix pencils to skew-Hamiltonian/Hamiltonian matrix pencils of larger dimension to get more reliable and accurate results. We also consider discrete-time systems, apply the extension strategy to the arising symplectic matrix pencils and transform these to more convenient structures in order to apply structure-exploiting eigenvalue solvers to them. We also investige a new structure-preserving method for the computation of the eigenvalues of skew-Hamiltonian/Hamiltonian matrix pencils and use this to increase the accuracy of the computed eigenvalues even more. In particular we ensure the reliability of the $L_\infty$-norm algorithm by this new eigenvalue solver. Finally we describe the implementation of the algorithms in Fortran and test them using two real-world examples.
172

Male Reproductive Strategies in Verreaux's Sifaka (Propithecus verreauxi) / Reproduktionsstrategien der Männchen beim Larven-Sifaka (Propithecus verreauxi)

Mass, Vanessa 23 April 2009 (has links)
No description available.
173

Costs of mate-guarding in wild male long-tailed macaques (Macaca fascicularis)

Girard-Buttoz, Cédric 28 October 2013 (has links)
In vielen promisken Paarungssystemen konkurrieren die Männchen einer Gruppe um den Zugang zu fertilen Weibchen. Um es Rivalen zu erschweren, haben sie verschiedene Paarungstaktiken entwickelt. Eine von Vertebraten und Invertebraten oft genutzte Strategie ist das „mate-guarding“. Hier bewacht ein hochrangiges Männchen das fertile Weibchen indem es ständig in dessen Nähe bleibt, wodurch es den Zugang der anderen Männchen stark minimiert. Durch diese Monopolisierung des Weibchens erhöht ein Männchen seinen Reproduktionserfolg und damit zusätzlich die Wahrscheinlichkeit der Vaterschaft. Diese für das Männchen gewinnbringende Strategie birgt jedoch auch energetische Kosten. Solche negativen Verknüpfungen zwischen Kosten und Gewinn, sogenannte „trade-offs“, beeinflussen den Fortpflanzungserfolg ebenso wie die Körperkondition und die Überlebenschance eines Männchens. Haben solche Kosten beispielsweise eine Verschlechterung der körperlichen Verfassung zur Folge, kann sich das negativ auf die Fähigkeiten der Männchen, ein Weibchen zu monopolisieren, auswirken und damit den Vaterschaftserfolg der Männchen mindern. Die mit solch einer Paarungstaktik wie dem „mate-guarding“ einhergehenden Kosten könnten sich auch auf die Entstehung von Strategien zur Partnerwahl bei den Männchen auswirken: Männchen sollten ihre Energie vor allem auf die Reproduktion mit den fittesten Weibchen aufwenden. Um die grundlegenden Faktoren der Partnerwahl sowie die Verteilung des Reproduktionserfolges unter den Männchen („reproductive skew“) besser zu verstehen, müssen die bei der Monopolisierung des Weibchens entstehenden Kosten quantifiziert werden. Primaten sind ein interessantes Taxa um diese Fragen zu untersuchen, da viele Arten in stabilen Mehr-Männchen-Gruppen leben und „mate-guarding“ eine vorteilhafte Taktik ist, die oft von hochrangigen Männchen angewandt wird. Allerdings haben sich bisherige Studien an Primaten auf die Quantifizierung der Futterkosten beschränkt und die Ergebnisse sind bisher sehr widersprüchlich. Unser Verständnis dieser Kosten wird weiterhin durch das Fehlen eines zuverlässigen, nicht-invasiven physiologischen Markers, der den energetischen Zustand von Nicht-Menschenaffen misst, beeinträchtigt. Das Hauptziel dieser Arbeit war es daher, die Kosten des „mate-guarding“ in einer Primatenart, die in Mehr-Männchen-Gruppen lebt wie die Javaneraffen (Macaca fascicularis), zu quantifizieren. Bisherige Ergebnisse zeigen, dass die Alpha-Männchen dieser Primatenart ihre Weibchen weniger monopolisieren als das „Priority of Access-Model“ vorhersagt. Der Monopolisierungserfolg scheint demnach durch die Kosten, die den Männchen durch das „mate-guarding“ entstehen, limitiert zu sein. In Studie 1 der vorliegenden Arbeit habe ich die Eignung von Urin C-Peptiden (UCP, ein Nebenprodukt der Insulinproduktion), als Marker für den Energiestatus von Makakenmännchen, evaluiert. In Studie 2 und 3 quantifizierte ich die energetischen, physiologischen und physischen (z.B. Aggression) Kosten des „mate-guardings“. In der vierten Studie untersuchte ich den Einfluss der Qualität der Weibchen auf die Kosten der des „mate-guardings“ und die Investition der Männchen in dieses Verhalten. Als erstes betrachtete ich den Zusammenhang zwischen den UCP Werten und Indikatoren für den Zustand der körperlichen Verfassung bei frei- und in Gefangenschaft lebender Makaken, um UCP als zuverlässigen Marker für Energiestatus (Studie 1) zu validieren. Die UCP Level waren positiv korreliert mit dem BMI (Body-Mass-Index) sowie mit dem Fettgehalt einer Hautfalte. In einem Experiment, bei dem die Futterzufuhr reduziert wurde, stellte sich heraus, dass UCP Level mit Änderungen des BMI und der geminderten Futterzufuhr kovariiert. Demzufolge ist UCP ein nützlicher Marker um nicht-invasiv intra- und interindividuelle Veränderungen der Körperkondition und des Ernährungszustandes zu ermitteln. Für die weitere Fragestellung beobachtete ich drei freilebende Javaneraffengruppen während zwei Paarungsperioden, in Ketambe, Gunung Leuser National Park in Indonesien. Um ein möglichst umfassendes Bild der potentiellen Kosten des „mate-guarding“ bereitzustellen, kombinierte ich zum einen meine durchgeführten Verhaltensbeobachtungen der Männchen, den Aufenthalt der Männchen in den Bäumen und sexuelle Interaktionen der Männchen mit den Weibchen. Zum anderen ermittelte ich GPS-Daten der Wanderungsdistanz, non-invasive Indikatoren für physiologischen Stress (faecal glucocorticoid, fGC), den Energiestatus (UCP) und bewertete die Verfügbarkeit von Früchten. Insgesamt konnten 2088 Fokusstunden, 331 Urin- und 771 Kotproben gesammelt und analysiert werden. Zudem wurden jeden Monat 360 Fruchtbäume begutachtet. In Studie 2 konnte ich zeigen, dass „mate-guarding“ die Parameter der Energieaufnahme und des Energieverbrauches reduziert. Dies hatte jedoch keine signifikanten Auswirkungen auf den gesamten Energiestatus (UCP Level) eines Männchens. Dieses Ergebnis weist auf ein ausbalanciertes Verhältnis von Energieaufnahme und Energieverbrauch der Männchen während des „mate-guardings“ hin. In Studie 3 konnte ich nachweisen, dass die Männchen während des „mate-guardings“, höhere fGC Werte aufwiesen. Dieser Wert wurde jedoch durch die Zeit, die Männchen in Vigilanz investieren, moduliert. „Mate-guarding” erhöhte einerseits die Vigilanzzeit eines Männchens und andererseits auch die Aggressionsrate der Männchen. Alpha-Männchen waren das ganze Jahr über gestresster als andere Männchen, unabhängig von Paarungskonkurrenz. Dies suggeriert, dass erhöhte Glucocorticoidlevel während des „mate-guarding“ den Männchen helfen ihre energetische Homöostase aufrechtzuerhalten, jedoch könnte dies Langzeitkosten darstellen, die bei lang anhaltender Belastung zu chronischem Stress führen können. Die Kombination dieser physiologischer Kosten und dem Verletzungsrisiko, dass mit Aggressionen einhergeht, könnte die Möglichkeit eines Alphamännchens ein Weibchen zu monopolisieren minimieren und damit auch Einfluss auf die Verteilung des Reproduktionserfolges der Männchen in einer Gruppe haben. In Studie 4 konnte ich zeigen, dass männliche Javaneraffen einige der Kosten des „mate-guarding“ reduzieren können indem sie gezielt Weibchen mit hohem reproduktiven Wert bewachen, da sie dann geringere fGC Werte haben. Darüber hinaus passten Männchen ihre Investition in „mate-guarding“ an, indem sie aufmerksamer und aggressiver waren wenn sie hochrangige Weibchen oder Weibchen mit denen sie starke Bindungen formten, bewachten. Diese Ergebnisse bestätigen, dass Männchen nicht nur hochwertige Weibchen auswählen, sondern diese auch länger und besser monopolisieren. In meiner Arbeit konnte ich die Kosten, die „mate-guarding“ für die Männchen einer Primatenart mit sich bringt, aufzeigen und hervorheben wie diese Kosten die Verteilung des Reproduktionserfolges unter den Männchen in der Gruppe beeinflusst. Auf Grundlage meiner Ergebnisse schlage ich vor, dass männliche Javaneraffen eine „unvollständige Weibchenmonopolisierungs-Strategie“ entwickelt haben, bei der sie die Kosten des „mate-guarding“ reduzieren indem sie Weibchen selektiv nach deren Reproduktionsqualität wählen und Weibchen mit geringerer Qualität weniger gründlich monopolisieren. Diese unvollständige Weibchenmonopolisierung könnte eine entscheidende Komponente des Energiemanagements von Alphamännchen sein, die ihnen erlaubt ganzjährig adäquat auf versuchte Rangübernahmen zu reagieren und somit ihre Amtszeit zu verlängern und die damit einhergehenden Fitnessvorteile zu erhalten. Beim Vergleich meiner Ergebnisse mit anderen Säugetier-Taxa, diskutiere ich in meiner Arbeit weiterhin die Beziehung zwischen den Kosten des „mate-guarding“ und der Verteilung des Reproduktionserfolges der Männchen in der Gruppe, die durch 1) reproduktive Saisonalität, 2) Energie-Management-Strategien der Männchen, 3) Errungenschaft eines hohen Ranges in der Gruppe und 4) der Sozialstruktur, moduliert sein kann. Zukünftige Studien, die die Kosten der Paarungstaktiken der Männchen untersuchen, sollten die Komplexität des Reproduktionsaufwandes, den Männchen investieren, bedenken. Diese Investitionen scheinen nicht ausschließlich auf die reproduktive Phase im Jahr beschränkt zu sein, sondern können sich über das ganze Jahr verteilen und spiegeln sich in Form der Konkurrenz zwischen Männchen in Bezug auf Rangstatus und sozialen Interaktionen wider.
174

Design and prototyping of temperature resilient clock distribution networks

Natu, Nitish Umesh 22 May 2014 (has links)
Clock Distribution Networks play a vital role in performance and reliability of a system. However, temperature gradients observed in 3D ICs hamper the functionality of CDNs in terms of varying skew and propagation delay. This thesis presents two compensation techniques, Adaptive Voltage and Controllable Delay, to overcome these problems. The compensation methods are validated using a FPGA-based test vehicle. Modification in traditional buffer design are also presented and the performance as well as the area and power overhead of both the implementations is compared.
175

Determination Of Network Delay Distribution Over The Internet

Karakas, Mehmet 01 December 2003 (has links) (PDF)
The rapid growth of the Internet and the proliferation of its new applications pose a serious challenge in network performance management and monitoring. The current Internet has no mechanism for providing feedback on network congestion to the end-systems at the IP layer. For applications and their end hosts, end-to-end measurements may be the only way of measuring network performance. Understanding the packet delay and loss behavior of the Internet is important for proper design of network algorithms such as routing and flow control algorithms, for the dimensioning of buffers and link capacity, and for choosing parameters in simulation and analytic studies. In this thesis, round trip time (RTT), one-way network delay and packet loss in the Internet are measured at different times of the day, using a Voice over IP (VoIP) device. The effect of clock skew on one-way network delay measurements is eliminated by a Linear Programming algorithm, implemented in MATLAB. Distributions of one-way network delay and RTT in the Internet are determined. It is observed that delay distribution has a gamma-like shape with heavy tail. It is tried to model delay distribution with gamma, lognormal and Weibull distributions. It is observed that most of the packet losses in the Internet are single packet losses. The effect of firewall on delay measurements is also observed.
176

O desvio reprodutivo se correlaciona positivamente com o parentesco genético e o sistema de acasalamento? Euglossa cordata (Hymenoptera, Apidae, Euglossini) como estudo de caso

Freiria, Gabriele Antico 26 June 2015 (has links)
Submitted by Daniele Amaral (daniee_ni@hotmail.com) on 2016-09-14T20:10:01Z No. of bitstreams: 1 TeseGAF.pdf: 2448190 bytes, checksum: 08c41156c28e3e99e6284a8c18acb2d2 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-16T19:49:11Z (GMT) No. of bitstreams: 1 TeseGAF.pdf: 2448190 bytes, checksum: 08c41156c28e3e99e6284a8c18acb2d2 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-16T19:49:17Z (GMT) No. of bitstreams: 1 TeseGAF.pdf: 2448190 bytes, checksum: 08c41156c28e3e99e6284a8c18acb2d2 (MD5) / Made available in DSpace on 2016-09-16T19:49:22Z (GMT). No. of bitstreams: 1 TeseGAF.pdf: 2448190 bytes, checksum: 08c41156c28e3e99e6284a8c18acb2d2 (MD5) Previous issue date: 2015-06-26 / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / The orchid bees have always been considered promising for studies of the evolution of social behavior and reproductive division of labor. This is due to the fact that the orchid bees are the only members not clearly eusocial in corbiculate clade. In this study, the correlation between genetic relatedness and reproductive skew was investigated in the genus Euglossa, one of the five genus of Euglossini tribe, in order to support the understanding of the evolution of reproductive division of labor in Euglossini. Among the five genera of Euglossini, Euglossa is the most suitable for studies on the evolution of reproductive division of labor, because in this genus can be found solitary, communal and primitively eusocial species. Twelve nests of Euglossa cordata (Linnaeus, 1758) were analyzed. In this species, the foundation process is solitary, but small colonies are formed during nest reactivations, with females establishing different types of associations, in which their relatedness differs. During reactivations, one of the females performs oophagy and then oviposition in brood cells previously operculated by another females, consequently leaving a larger number of offspring. The behavioral analysis associated with relatedness estimates, made through the use of microsatellite markers, indicated that the dominant female perform oophagy followed by oviposition in all operculated cells by the subordinated bees. The reproductive skew was complete, regardless of the type of association established between the interacting females, and so reproductive skew and genetic relatedness was not associated in this study. Furthermore, the nesting biology of Euglossa was also studied, given the information scarcity about the nesting process in most species already described for this genus. Using the technique of trap nests, 43 nests of Euglossa were sampled. This study provides new information on the nesting biology of this genus, especially with regard to natural enemies associated with nests. / As abelhas da tribo Euglossini sempre foram consideradas promissoras para estudos sobre evolução do comportamento social e divisão de trabalho reprodutivo. Isto decorre do fato dos Euglossini serem os únicos membros não claramente eussociais entre as abelhas corbiculadas. Neste estudo, foi investigada a correlação entre parentesco genético e desvio reprodutivo em uma espécie do gênero Euglossa, um dos cinco gêneros de Euglossini, visando subsidiar o entendimento da evolução da divisão de trabalho reprodutivo na tribo. Dentre os cinco gêneros de Euglossini, Euglossa é o mais indicado para estudos sobre a evolução de divisão de trabalho reprodutivo, pois neste gênero são encontradas espécies solitárias, comunais e primitivamente eussociais. Foram analisados doze ninhos da espécie Euglossa cordata (Linnaeus, 1758). Nesta espécie, o processo de fundação dos ninhos é solitário, mas pequenas colônias são formadas durante as reativações, com as fêmeas estabelecendo diferentes tipos de associações, nas quais o parentesco entre elas é distinto. Durante as reativações, uma das fêmeas realiza oofagias, seguidas de oviposições em células anteriormente operculadas por outras fêmeas, deixando um maior número de descendentes. As análises de comportamento, associadas a estimativas de parentesco, realizadas mediante o uso de marcadores microssatélites, indicam que as fêmeas dominantes realizam oofagia, seguida de oviposição, em todas as células operculadas pelas subordinadas. O desvio reprodutivo foi completo, independentemente do tipo de associação formada entre as fêmeas reativadoras e, portanto, não houve correlação entre parentesco genético e desvio reprodutivo. Além disso, a biologia de nidificação em espécies de Euglossa foi estudada, dada a escassez de informações sobre os hábitos e características do processo de nidificação na maioria das espécies já descritas. Utilizando a técnica de ninhos-armadilha, foram obtidos 44 ninhos de Euglossa. O estudo traz novas informações sobre a biologia de nidificação no gênero, principalmente no que diz respeito aos inimigos naturais associados aos ninhos.
177

Segmenta??o fuzzy de objetos tridimensionais com propriedades texturais / Segmentation of three-dimensional objects with textural propertie

Silva Neto, Jos? Francisco da 25 September 2014 (has links)
Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2016-01-26T17:46:31Z No. of bitstreams: 1 JoseFranciscoDaSilvaNeto_DISSERT.pdf: 5950864 bytes, checksum: 5306cd9802b9aa1c09288d75c32ccbe2 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2016-01-28T18:00:31Z (GMT) No. of bitstreams: 1 JoseFranciscoDaSilvaNeto_DISSERT.pdf: 5950864 bytes, checksum: 5306cd9802b9aa1c09288d75c32ccbe2 (MD5) / Made available in DSpace on 2016-01-28T18:00:31Z (GMT). No. of bitstreams: 1 JoseFranciscoDaSilvaNeto_DISSERT.pdf: 5950864 bytes, checksum: 5306cd9802b9aa1c09288d75c32ccbe2 (MD5) Previous issue date: 2014-09-25 / Segmenta??o digital de imagens ? o processo de atribuir r?tulos distintos a diferentes objetos em uma imagem digital, e o algoritmo de segmenta??o fuzzy tem sido utilizado com sucesso na segmenta??o de imagens de diversas modalidades. Contudo, o algoritmo tradicional de segmenta??o fuzzy falha ao segmentar objetos que s?o caracterizados por texturas cujos padr?es n?o podem ser descritos adequadamente por simples estat?sticas computadas sobre uma ?rea restrita. Neste trabalho apresentamos uma extens?o do algoritmo de segmenta??o fuzzy que realiza segmenta??o de texturas empregando fun??es de afinidade adaptativas e o estendemos a imagens tridimensionais. Fun??es de afinidade adaptativas mudam o tamanho da ?rea em que s?o calculados os descritores da textura de acordo com as caracter?sticas da textura processada, enquanto imagens tridimensionais podem ser descritas como um conjunto finito de imagens bidimensionais. O algoritmo ent?o segmenta o volume com uma ?rea apropriada calculada para cada textura, tornando poss?vel obter boas estimativas dos volumes reais das estruturas alvo do processo de segmenta??o. Experimentos ser?o realizados com dados sint?ticos e reais obtidos no estudo de segmenta??o de tumores cerebrais em imagens m?dicas adquiridas atrav?s de exames de Resson?ncia Magn?tica / Digital image segmentation is the process of assigning distinct labels to different objects in a digital image, and the fuzzy segmentation algorithm has been used successfully in the segmentation of images from several modalities. However, the traditional fuzzy segmentation algorithm fails to segment objects that are characterized by textures whose patterns cannot be successfully described by simple statistics computed over a very restricted area. In this paper we present an extension of the fuzzy segmentation algorithm that achieves the segmentation of textures by employing adaptive affinity functions as long as we extend the algorithm to tridimensional images. The adaptive affinity functions change the size of the area where they compute the texture descriptors, according to the characteristics of the texture being processed, while three dimensional images can be described as a finite set of two-dimensional images. The algorithm then segments the volume image with an appropriate calculation area for each texture, making it possible to produce good estimates of actual volumes of the target structures of the segmentation process. We will perform experiments with synthetic and real data in applications such as segmentation of medical imaging obtained from magnetic rosonance
178

Análise de fadiga em pontes rodoviárias de concreto armado / Fatigue analysis of reinforced concrete road bridges

Antonio Jorge Figueiredo Ahi 21 August 2009 (has links)
As pontes rodoviárias de concreto armado estão sujeitas às ações dinâmicas variáveis devido ao tráfego de veículos sobre o tabuleiro. Estas ações dinâmicas podem gerar o surgimento das fraturas ou mesmo a sua propagação na estrutura. A correta consideração destes aspectos objetivou o desenvolvimento de um estudo, de forma a avaliar os esforços do tráfego de veículos pesados sobre o tabuleiro. As técnicas para a contagem de ciclos de esforços e a aplicação das regras de dano acumulado foram analisadas através das curvas S-N de diversas normas estudadas. A ponte rodoviária investigada é constituída por quatro vigas longitudinais, três transversinas e por um tabuleiro de concreto armado. O modelo computacional, desenvolvido para a análise dinâmica da ponte, foi concebido com base no emprego de técnicas usuais de discretização através do método dos elementos finitos. O modelo estrutural da obra de arte rodoviária estudada foi simulado com base no emprego de elementos finitos sólidos tridimensionais. Os veículos são representados a partir de sistemas massa-mola-amortecedor. O tráfego dessas viaturas é considerado mediante a simulação de comboios semi-infinitos, deslocando-se com velocidade constante sobre o tabuleiro da ponte. As conclusões deste trabalho versam acerca da vida útil de serviço dos elementos estruturais de pontes rodoviárias de concreto armado submetidas às ações dinâmicas provenientes do tráfego de veículos pesados sobre o tabuleiro. / Reinforced concrete highway bridges are subjected to dynamic actions of variable magnitude due to vehicles crossing on the deck pavement. These dynamic actions can generate the nucleation of fractures or even their propagation on the structure. This way, the behaviour and reliability can be compromised and the bridge service life can be reduced. The stress cycles counting techniques and the cumulative damage rules application had been analyzed through S-N curves, based on an extensive revision of concrete bridges service life and theoretical fatigue aspects in reinforced concrete structures, as well as considering the recommended procedures used for the concrete structures main codes. A reinforced concrete skew bridge with spanning 16.5m was investigated in this work. The computational model, developed for the composite bridge dynamic analysis, adopted the usual mesh refinement techniques present in finite element method simulations implemented in the ANSYS program. These viactures are simulated by spring-mass models and the transit are considered by means of semi infinity convoys, displaced with constant velocity across the deck pavement. The proposed analysis methodology and the procedures presented in the design codes were applied, in order to evaluate the bridge fatigue determining the structure service life. The conclusions of this paper have the objective of alerting structural engineers to the possible distortions, associated to the reinforced concrete bridges service life when subjected to dynamic actions due to vehicles crossing on the deck pavement surface.
179

Diagnóstico de influência em modelos com erros na variável skew-normal/independente / Influence of diagnostic in models with errors in variable skew-normal/independent

Carvalho, Rignaldo Rodrigues 17 August 2018 (has links)
Orientadores: Victor Hugo Lachos Dávila, Filidor Edilfonso Vilca Labra / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática, Estatística e Computação Científica / Made available in DSpace on 2018-08-17T09:37:18Z (GMT). No. of bitstreams: 1 Carvalho_RignaldoRodrigues_M.pdf: 1849605 bytes, checksum: 07ea5638a2dbfa2227f9a949d4723bbf (MD5) Previous issue date: 2010 / Resumo: O modelo de medição de Barnett é frequentemente usado para comparar vários instrumentos de medição. é comum assumir que os termos aleatórios têm uma distribuição normal. Entretanto, tal suposição faz a inferência vulnerável a observações atípicas por outro lado distribuições de misturas de escala skew-normal tem sido uma interessante alternativa para produzir estimativas robustas tendo a elegância e simplicidade da teoria da máxima verossimilhança. Nós usamos resultados de Lachos et al. (2008) para obter a estimação dos parâmetros via máxima verossimilhança, baseada no algoritmo EM, o qual rende expressões de forma fechada para as equações no passo M. Em seguida desenvolvemos o método de influência local de Zhu e Lee (2001) para avaliar os aspectos de estimação dos parâmetros sob alguns esquemas de perturbação. Os resultados obtidos são aplicados a conjuntos de dados bastante estudados na literatura, ilustrando a utilidade da metodologia proposta / Abstract: The Barnett measurement model is frequently used to comparing several measuring devices. It is common to assume that the random terms have a normal distribution. However, such assumption makes the inference vulnerable to outlying observations whereas scale mixtures of skew-normal distributions have been an interesting alternative to produce robust estimates keeping the elegancy and simplicity of the maximum likelihood theory. We used results in Lachos et al. (2008) for obtaining parameter estimation via maximum likelihood, based on the EM-algorithm, which yields closed form expressions for the equations in the M-step. Then we developed the local influence method to assessing the robustness aspects of these parameter estimates under some usual perturbation schemes. Results obtained for one real data set are reported, illustrating the usefulness of the proposed methodology / Mestrado / Métodos Estatísticos / Mestre em Estatística
180

Misturas finitas de misturas de escala skew-normal / Mixtures modelling using scale mixtures of skew-normal distribution

Basso, Rodrigo Marreiro 03 December 2009 (has links)
Orientador: Victor Hugo Lachos Davila / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Computação Cientifica / Made available in DSpace on 2018-08-13T07:03:11Z (GMT). No. of bitstreams: 1 Basso_RodrigoMarreiro_M.pdf: 3130269 bytes, checksum: 85e95beb812a4ec069f39f8b9c79681a (MD5) Previous issue date: 2009 / Resumo: Nesse trabalho será considerada uma classe flexível de modelos usando misturas finitas de distribuições da classe de misturas de escala skew-normal. O algoritmo EM é empregado para se obter estimativas de máxima verossimilhança de maneira iterativa, sendo discutido com maior ênfase para misturas de distribuições skew-normal, skew-t, skew-slash e skew-normal contaminada. Também será apresentado um método geral para aproximar a matrix de covariância assintótica das estimativas de máxima verossimilhança. Resultados obtidos da análise de quatro conjuntos de dados reais ilustram a aplicabilidade da metodologia proposta / Abstract: In this work we consider a flexible class of models using finite mixtures of multivariate scale mixtures of skew-normal distributions. An EM-type algorithm is employed for iteratively computing maximum likelihood estimates and this is discussed with emphasis on finite mixtures of skew-normal, skew-t, skew-slash and skew-contaminated normal distributions. A general information-based method for approximating the asymptotic covariance matrix of the maximum likelihood estimates is also presented. Results obtained from the analysis of four real data sets are reported illustrating the usefulness of the proposed methodology / Mestrado / Mestre em Estatística

Page generated in 0.1083 seconds