Basiléia II e exigência de capital para risco de crédito dos bancos no Brasil

Com a implementação do Acordo de Basiléia II no Brasil, os grandes conglomerados bancários poderão utilizar o chamado modelo IRB (Internal Ratings Based) para cômputo da parcela de risco de crédito da exigência de capital. O objetivo deste trabalho é mensurar a diferença entre o capital mínimo exigido (e, conseqüentemente, do Índice de Basiléia) calculado pela abordagem IRB em relação à regulamentação atual. Para isso, foram estimadas probabilidades de inadimplência (PD) utilizando matrizes de transição construídas a partir dos dados da Central de Risco de Crédito (SCR) do Banco Central do Brasil. Os resultados indicam aumento da exigência de capital, ao contrário do ocorrido nos países do G-10

As competências centrais da Dell : uma avaliação sobre sua aplicabilidade em serviços

O propósito principal deste trabalho foi avaliar como as competências centrais de Dell Inc., que formam a base de sua vantagem competitiva, contribuem ou não para a iniciativa da empresa de ingressar no segmento de prestação de serviços, na indústria da Tecnologia da Informação. Para identificar as competências centrais da Dell Inc. foram utilizados os critérios propostos na teoria Resource Based Model.

Ensaios sobre inflação e política monetária

Esta tese tem por objetivo principal o estudo da relação entre atividade econômica, inflação e política monetária no tocante a três aspectos importantes. O primeiro, a perspectiva histórica da evolução da relação entre atividade e inflação no pensamento econômico. O segundo, a análise da dinâmica inflacionária utilizando um modelo com fundamentação microeconômica, no caso a curva de Phillips Novo-Keynesiana, com uma aplicação ao caso brasileiro. O terceiro, a avaliação da eficiência dos mecanismos de sinalização de política monetária utilizados pelo Banco Central no Brasil com base nos movimentos na estrutura a termo da taxa de juros com a mudança da meta da Selic. O elemento central que une estes ensaios é a necessidade do formulador de política econômica compreender o impacto significativo das ações de política monetária na definição do curso de curto prazo da economia real para atingir seus objetivos de aliar crescimento econômico com estabilidade de preços. Os resultados destes ensaios indicam que o modelo Novo-Keynesiano, resultado de um longo desenvolvimento na análise econômica, constitui-se numa ferramenta valiosa para estudar a relação entre atividade e inflação. Uma variante deste modelo foi empregada para estudar com relativo sucesso a dinâmica inflacionária no Brasil, obtendo valores para rigidez da economia próximos ao comportamento observado em pesquisas de campo. Finalmente, foi aliviada a previsibilidade das ações do Banco Central para avaliar o estágio atual de desenvolvimento do sistema de metas no Brasil, através da reação da estrutura a termo de juros às mudanças na meta da taxa básica (Selic). Os resultados indicam que comparando o período de 2003 a 2008 com 2000 a 2003, verificamos que os resultados apontam para o aumento da previsibilidade das decisões do Banco Central. Este fato pode ser explicado por alguns fatores: o aprendizado do público sobre o comportamento do Banco Central; a menor volatilidade econômica no cenário econômico e o aperfeiçoamento dos mecanismos de sinalização e da própria operação do sistema de metas. Comparando-se o efeito surpresa no Brasil com aqueles obtidos por países que promoveram mudanças significativas para aumentar a transparência da política monetária no período de 1990 a 1997, observa-se que o efeito surpresa no Brasil nas taxas de curto prazo reduziu-se significativamente. No período de 2000 a 2003, o efeito surpresa era superior aos de EUA, Alemanha e Reino Unido e era da mesma ordem de grandeza da Itália. No período de 2003 a 2008, o efeito surpresa no Brasil está próximo dos valores dos EUA e Alemanha e inferiores aos da Itália e Reino Unido.

Modelo interno de risco de crédito de Basiléia II: possíveis impactos no capital mínimo exigido dos bancos

Com a implementação do Acordo de Basiléia II no Brasil, os grandes conglomerados bancários poderão utilizar o chamado modelo IRB (Internal Ratings Based) para cômputo da parcela de risco de crédito da exigência de capital. O objetivo desta dissertação é mensurar a diferença entre o capital mínimo exigido (e, conseqüentemente, do Índice de Basiléia) calculado pela abordagem IRB em relação à regulamentação atual. Para isso, foram estimadas probabilidades de inadimplência (PD) utilizando matrizes de transição construídas a partir dos dados da Central de Risco de Crédito (SCR) do Banco Central do Brasil. Os resultados indicam aumento da exigência de capital, ao contrário do ocorrido nos países do G-10.

Pricing and hedging of oil futures : a unifying approach

We develop and empirically test a continuous time equilibrium model for the pricing of oil futures. The model provides a link between no-arbitrage models and expectation oriented models. It highlights the role of inventories for the identification of different pricing regimes. In an empirical study the hedging performance of our model is compared with five other one- and two-factor pricing models. The hedging problem considered is related to Metallgesellschaft´s strategy to hedge long-term forward commitments with short-term futures. The results show that the downside risk distribution of our inventory based model stochastically dominates those of the other models.

Crowdfunding no Brasil: uma análise sobre as motivações de quem participa

Este trabalho procurou investigar as motivações para a participação de apoiadores no modelo de recompensa de crowdfunding no Brasil, sob a luz dos aspectos encontrados na pesquisa americana de Gerber e Hui (2014). Como a participação nesse modelo é voluntária, entendeu- se ser importante compreender os motivos que levam pessoas a apoiarem projetos. Acredita- se que este trabalho tenha atingido o que foi por ele proposto, deixando contribuições em diversos sentidos. A fim de aprofundar o entendimento desse novo fenômeno social, apresentou-se uma pesquisa qualitativa fundamentada em um estudo de caso múltiplo, em que os apoiadores eram a unidade de análise, nas três maiores plataformas de crowdfunding do Brasil: Queremos, Catarse e Benfeitoria. E, como fonte de informações para esta metodologia, optou-se pelo método qualitativo de entrevistas em profundidade com os elementos da unidade de análise. Foram realizadas 11 entrevistas com apoiadores, sendo 06 homens e 05 mulheres. O trabalho também teve o objetivo de conhecer melhor o cenário nacional desse mercado, a partir de entrevistas em profundidade com os fundadores das plataformas e um heavy user (mais de 140 projetos apoiados) do modelo. Após a consolidação e análise dos dados obtidos, verificou-se a presença das motivações encontradas nos estudos de Gerber (GERBER e HUI, 2014), porém com algumas ressalvas quanto a motivação “Fazer parte de uma comunidade”, explicitada a seguir. A pesquisa qualitativa refinou substancialmente a compreensão do que motiva apoiadores a participar de crowdfunding, incluindo aspectos importantes que devem ser levados em consideração quanto a práticas do mercado. Ao final, as conclusões e implicações deste estudo foram detalhadamente apresentadas.

Efeito inseticida de vicilinas isoladas de sementes de erythrina velutina em condições de semi-campo para moscas das frutas (ceratitis capitata)

Chitin-binding vicilins from legume seeds (Erythrina velutina. Canavalia ensiformes and Phaseolus vulgares) were isolated by ammonium sulfate followed by affinity chromatography on a chitin column. Effect of these vicilins on female adults of Ceratitis capitata was examined by bioassay and in a semi-field assay model. Mechanism of action of the vicilins was determined by in vivo digestibility and chitin affinity. Among the tested vicilins, E. velutina when added to diet caused strong effect on mortality at 10% dose. This insecticidal property was tested in a semi-field assay which showed the same effect observed in laboratory conditions, where doses of 10% and 15% were lethal to female adults of C. capitata. These deleterious effects were not only associated to the binding to chitin structures present in peritrophic membrane, but principally to its low digestibility in the C. capitata digestive tract. This fact was confirmed because chiting binding proteins as WGA and the other tested vicilins were not toxic to female adults of C. capitata due susceptibility of these proteins to digestive enzymes of the insects. By other side EvV was more resistant to digestive enzymes, causing deleterious effects on female adults of C. capitata. These results showed that EvV may be part of the pest management programs or an alternative in plant improvement program in the population control of this fruticulture pest

Automação da drenagem no teste de produção convencional em tanque cilíndrico

This document aims to improve the quality of the production test in vertical tanks with free water drain pipes, through a device to control the draining system. This proposal consists of an interface detector close to the tank bottle and a control valve on the pipe-drain; they are attached to a remote supervisor system, which will be minimizing the human influence in the conclusion of the test result. And for more consciousness the work shows the importance of the wells production test in the attendance and diagnosis of the productive process, informing the large number of tests executed and problems of the procedure adopted in the field today. There are many possible sources of uncertainty in this kind of test as shown in the experiments realized in the field; the object prototype of this dissertation will be made in the field, based upon the definition of parameters and characteristics of the devices proposal. For a better definition of the draining process the action results of the assessment test are shown, especially changed some for the understand ing of the real process. It shows the proposal details and the configuration that will be used in the tank of Monte Alegre s field Production Station, explaining the interface detector kind and the control system. It is the base to a pilot project now in development, named as the new project classified in the status of the new technology and production improvement of PETROBRAS in Rio Grande do Norte and Ceará. This dissertation concludes that the automation of the conventional test with the draining system will bring benefits both economically as metrologically, because it reduces the uncertainty of the test procedures with free water draining, and also decreases the number of tests with problems

Redes sociais e a lógica do desvio comportamental em relação às normas de trabalho

This paper aim to check a hypothesis that assumes several behaviors related to social work norm´s obeying as a phenomenon that can be explained by actor´s social network structure and the rational choice processes related to the social norm inside that network, principally the payoff´s analysis received by the closest actors, or neighbors, at a social situation. Taking the sociological paradigm of rational action theory as a basis, the focus is on a debate about the logic of social norms, from Émile Durkheim´s method to Jon Elster´s theory, but also including social network analysis´s variables according to Robert Hanneman; and also Vilfredo Pareto´s constants related to human sociability, at the aim to detect elements that can help the scholars to develop an agent based model which could explain the sociological problem of deviance by a better way than the common sense´s view about morality and ethics at a social work environment

Dores e cores nas mal traçadas linhas dos devotos do Padre Cícero: as trocas linguísticas instauradas entre o discurso eclesial e o discurso epistolar dos romeiros

This study is concerned with analyzing the letters sent by the devotees tothe religious patriarch, Father Cicero, checking as is the relationship of linguistic exchanges between the pilgrims and the Father Cicero through the issuance of letters; and, between the Church and the pilgrims, through homilies uttered these devotees, focusing on the appeals contained therein, for, then, to define a parameter between the requirements contained in the letters and the religious discourse of the Church workers in the liturgical celebrations in view of social and religious demands of the market in question. Our purposes resided in understanding the causes that lead to the production of the letters by devotees, holding us in prayers of intercession to the Father Cicero. Finally, check and understand how is the process of inter - relationship between the writing of devotee and the clerical discourse, with regard to meeting this demand, noting especially the game force a religious field, based on postulates of Pierre Bourdieu (2008), to conceive the communicative act as linguistic exchanges, surpassing the sign and therefore decipherable speech character as well as the interactive strength of the wording and voice advocated by Paul Zumthor (2010) e Bakhtin (2006). Regarding the composition of the corpus, were determined, as the universe of research, letters of pilgrims, sent to Father Cicero, deposited in his own tomb, as well as in the Church of the Horto (Gethsemane) in Juazeiro do Norte-CE (Brazil); beyond religious sermons intended for pilgrims during the massin pilgrimage time

Structural damage detection by fuzzy clustering

The development of strategies for structural health monitoring (SHM) has become increasingly important because of the necessity of preventing undesirable damage. This paper describes an approach to this problem using vibration data. It involves a three-stage process: reduction of the time-series data using principle component analysis (PCA), the development of a data-based model using an auto-regressive moving average (ARMA) model using data from an undamaged structure, and the classification of whether or not the structure is damaged using a fuzzy clustering approach. The approach is applied to data from a benchmark structure from Los Alamos National Laboratory, USA. Two fuzzy clustering algorithms are compared: fuzzy c-means (FCM) and Gustafson-Kessel (GK) algorithms. It is shown that while both fuzzy clustering algorithms are effective, the GK algorithm marginally outperforms the FCM algorithm. (C) 2008 Elsevier Ltd. All rights reserved.

Uma estratégia dirigida a modelos e baseada em linguagem de descrição arquitetural para linhas de produtos de software

Model-oriented strategies have been used to facilitate products customization in the software products lines (SPL) context and to generate the source code of these derived products through variability management. Most of these strategies use an UML (Unified Modeling Language)-based model specification. Despite its wide application, the UML-based model specification has some limitations such as the fact that it is essentially graphic, presents deficiencies regarding the precise description of the system architecture semantic representation, and generates a large model, thus hampering the visualization and comprehension of the system elements. In contrast, architecture description languages (ADLs) provide graphic and textual support for the structural representation of architectural elements, their constraints and interactions. This thesis introduces ArchSPL-MDD, a model-driven strategy in which models are specified and configured by using the LightPL-ACME ADL. Such strategy is associated to a generic process with systematic activities that enable to automatically generate customized source code from the product model. ArchSPLMDD strategy integrates aspect-oriented software development (AOSD), modeldriven development (MDD) and SPL, thus enabling the explicit modeling as well as the modularization of variabilities and crosscutting concerns. The process is instantiated by the ArchSPL-MDD tool, which supports the specification of domain models (the focus of the development) in LightPL-ACME. The ArchSPL-MDD uses the Ginga Digital TV middleware as case study. In order to evaluate the efficiency, applicability, expressiveness, and complexity of the ArchSPL-MDD strategy, a controlled experiment was carried out in order to evaluate and compare the ArchSPL-MDD tool with the GingaForAll tool, which instantiates the process that is part of the GingaForAll UML-based strategy. Both tools were used for configuring the products of Ginga SPL and generating the product source code

Uma análise da aplicação do modelo de Rede Neural RePART em Comitês de classificadores

RePART (Reward/Punishment ART) is a neural model that constitutes a variation of the Fuzzy Artmap model. This network was proposed in order to minimize the inherent problems in the Artmap-based model, such as the proliferation of categories and misclassification. RePART makes use of additional mechanisms, such as an instance counting parameter, a reward/punishment process and a variable vigilance parameter. The instance counting parameter, for instance, aims to minimize the misclassification problem, which is a consequence of the sensitivity to the noises, frequently presents in Artmap-based models. On the other hand, the use of the variable vigilance parameter tries to smoouth out the category proliferation problem, which is inherent of Artmap-based models, decreasing the complexity of the net. RePART was originally proposed in order to minimize the aforementioned problems and it was shown to have better performance (higer accuracy and lower complexity) than Artmap-based models. This work proposes an investigation of the performance of the RePART model in classifier ensembles. Different sizes, learning strategies and structures will be used in this investigation. As a result of this investigation, it is aimed to define the main advantages and drawbacks of this model, when used as a component in classifier ensembles. This can provide a broader foundation for the use of RePART in other pattern recognition applications

X-ARM: um modelo de representação de artefatos de software

A great challenge of the Component Based Development is the creation of mechanisms to facilitate the finding of reusable assets that fulfill the requirements of a particular system under development. In this sense, some component repositories have been proposed in order to answer such a need. However, repositories need to represent the asset characteristics that can be taken into account by the consumers when choosing the more adequate assets for their needs. In such a context, the literature presents some models proposed to describe the asset characteristics, such as identification, classification, non-functional requirements, usage and deployment information and component interfaces. Nevertheless, the set of characteristics represented by those models is insufficient to describe information used before, during and after the asset acquisition. This information refers to negotiation, certification, change history, adopted development process, events, exceptions and so on. In order to overcome this gap, this work proposes an XML-based model to represent several characteristics, of different asset types, that may be employed in the component-based development. Besides representing metadata used by consumers, useful for asset discovering, acquisition and usage, this model, called X-ARM, also focus on helping asset developers activities. Since the proposed model represents an expressive amount of information, this work also presents a tool called X-Packager, developed with the goal of helping asset description with X-ARM

Edge detection and noise removal by use of a partial differential equation with automatic selection of parameters

This work deals with noise removal by the use of an edge preserving method whose parameters are automatically estimated, for any application, by simply providing information about the standard deviation noise level we wish to eliminate. The desired noiseless image u(x), in a Partial Differential Equation based model, can be viewed as the solution of an evolutionary differential equation u t(x) = F(u xx, u x, u, x, t) which means that the true solution will be reached when t ® ¥. In practical applications we should stop the time ''t'' at some moment during this evolutionary process. This work presents a sufficient condition, related to time t and to the standard deviation s of the noise we desire to remove, which gives a constant T such that u(x, T) is a good approximation of u(x). The approach here focused on edge preservation during the noise elimination process as its main characteristic. The balance between edge points and interior points is carried out by a function g which depends on the initial noisy image u(x, t0), the standard deviation of the noise we want to eliminate and a constant k. The k parameter estimation is also presented in this work therefore making, the proposed model automatic. The model's feasibility and the choice of the optimal time scale is evident through out the various experimental results.