The economic literature search for some decades to analyze and to obtain empirical findings on the relationship between the financial system and economic growth. This research aims to evaluate the effect of two financial variables have on the productivity of brazilian states, from 1991 to 2010. Data were extracted from the Central Bank of Brazil and IPEADATA / IBGE. Thus, the data will be evaluated by the econometric method of analysis panel and calculated by using three models grounded in the text of Firme e Freguglia (2013). The results show that there is a positive and significant relationship between savings and productivity, as well as between financing and productivity of the states. / A literatura econômica busca há algumas décadas analisar e obter resultados empíricos sobre a relação entre o sistema financeiro e o crescimento econômico. Essa pesquisa objetiva avaliar o efeito que duas variáveis financeiras exercem sobre a produtividade dos estados brasileiros, no período de 1991 a 2010. Os dados foram extraídos do Banco Central do Brasil e do IPEADATA/IBGE. Assim, os dados serão avaliados por meio do método econométrico de análise em painel e calculados através de três modelos embasados no texto de Firme e Freguglia (2013). Os resultados mostram que há uma relação positiva e significativa entre a poupança e a produtividade, como também, entre o financiamento e a produtividade dos estados.

This work presents the use of analytical techniques, conventional and advanced for the chemical characterization of four samples of blends crude oils from the PETROBRAS station, Oiteirinhos II (Carmópolis-SE). These samples, which have different exploration characteristics, were identified as CEOL 1, CEOL 2, CEOL 3 and CEOL 4. After the collection in the field and subsequently separation of water residue through centrifugation at 2500 rpm, the blends were characterized by density, asphaltenes content, fractionation SARA (saturates, aromatics, resins and asphaltenes) and by the application of micropyrolisys off-line. The advanced characterization to elucidate the chemical relationship between the samples were made using the Thermogravimetry (TGA), Infrared Spectroscopy (IR), Spectroscopy in Ultraviolet (UV), Matrix Assisted Laser Desorption Ionization - Time of Flight - Mass Spectrometry (MALDI-TOF-MS), Gas Chromatography Mass Spectrometry (GC/MS), Gas Chromatography with Flame Ionization Detector (GC-FID) and Comprehensive Two-Dimensional Gas Chromatography (GC x GC). The blends CEOL 3 presented the highest density (0.9737 g mL-1), confirming the results of MALDI-TOF-MS, that shows predominantly high molecular mass compounds in the composition. The SARA fractionation indicates the Saturated fraction as the largest fraction in all oils (> 50%). The analysis of TGA explained similar bands of mass loss in all blends and their asphaltenes. The characterization by IR showed bands related to the hydrocarbon composition and the UV showed the presence of mono, di and tri aromatics structures in the blends as well in their respective asphaltenes. The chromatographic analysis (GC/MS, GC-FID and GC x GC) showed that all blends have a paraffinic profile, with the predominance of n-alkanes of low molecular weight (n-C10 to n-C40) and characteristics of oxide depositional environment (Pr/Ph>1) and lacustrine origin (CPI>1). The analysis by GC/MS showed that the distribution area relating to the free n-alkanes in the CEOL 3 blends is approximately 30% lower than the other samples, and that after the micropyrolysis process these areas become approximately the same. For biomarkers, GC/MS was possible to identify the majority presence of tri- and tetracyclic terpanes (m/z 191) and regular steranes (m/z 217) as the major constituents of the blends, pointing also to the release of occluded biomarkers in asphaltenic structures after the micropyrolysis process. The identification and subsequent correlation between biomarkers indicated that the blends crude oils studied have partial biodegradation characteristics, with a blend of mature and immature oils from terrigenous and/or algal organic matter. / Este trabalho apresenta o emprego de diversas técnicas analíticas, convencionais e avançadas, para a caracterização química de quatro blendas de óleos brutos provenientes da estação da Petrobras de Oiteirinhos II (Carmópolis-SE). Estas amostras, que apresentam diferentes características de exploração, foram identificadas como CEOL 1, CEOL 2, CEOL 3 e CEOL 4. Após a coleta em campo e posterior separação dos resíduos de água por centrifugação a 2500 rpm, as blendas foram caracterizados a partir da determinação da densidade, do teor de asfaltenos, fracionamento SARA (Saturados, Aromáticos, Resinas e Asfaltenos) e aplicação da micropirólise off-line. A caracterização avançada, aplicada para elucidar a relação química entre as amostras, foram realizadas utilizando as técnicas de Termogravimetria (TGA), Espectroscopia no Infravermelho (IV), Espectroscopia no Ultravioleta (UV), Espectrometria de Massa por Tempo de voo com Desorção-Ionização a Laser Assistida por Matriz (MALDI-TOF-MS), Cromatografia Gasosa acoplada a Espectrometria de Massas (CG/EM), Cromatografia Gasosa com Detector de Ionização de Chama (CG-DIC) e Cromatografia Gasosa Bidimensional Abrangente (CG x CG). A blenda CEOL 3 apresentou a maior densidade (0,9737 g mL-1) corroborando com os resultados de MALDI-TOF-MS, mostrando a predominância de compostos de alta massa molecular em sua composição. O fracionamento SARA aponta a fração de Saturados como a maior fração em todos as blendas (>50%). A análise de TGA expôs faixas de perdas de massas semelhantes em todas as amostras e seus asfaltenos. A caracterização por IV apresentou estiramentos referentes à composição de hidrocarbonetos e o UV a presença de estruturas mono, di e tri aromáticos tanto nas blendas como em seus respectivos asfaltenos. As análises cromatográficas (CG/EM, CG-DIC e CG x CG) mostraram que todos as blendas possuem um perfil parafínico, com predominância de n-alcanos de baixo peso molecular (n-C10 a n-C40) com características de ambiente deposicional óxido (Pr/Ft>1) de origem lacustre (CPI>1). A análise por CG/EM mostrou ainda que a distribuição em área referente a n-alcanos livres presentes na blenda CEOL 3 é cerca de 30% menor do que as outras amostras, e que após o processo de micropirólise essas áreas passam a ser aproximadamente iguais. Para biomarcadores, por CG/EM foi possível identificar a presença majoritária de terpanos tri e tetracíclicos (m/z 191) e esteranos regulares (m/z 217), como constituintes majoritários das blendas em estudo, apontando ainda, a liberação de biomarcadores ocluídos nas estruturas asfaltênicas após o processo de micropirólise. A identificação e posterior correlação entre os biomarcadores indicaram que as blendas de óleos brutos possuem características de biodegradação parcial, com uma mistura de muito e pouco evoluídos termicamente, provenientes de matéria orgânicas terrígenas e/ou algas.

This paper examines the hypothesis of GDP per capita convergence in the northeastern states, the period 1995-2009, with a focus on human capital. The reason to analyze the Northeast only and not all regions of Brazil is to retain the condition of homogeneity, necessary to analyze convergence issues. Using theories and models of economic growth, we tested the hypothesis of convergence ß- (absolute and conditional) convergence, and -o- during the period analyzed. Using Stata software version 12.1, the coefficients were estimated using panel data with fixed effects, whose main advantage is in capturing behavior changes of the indicators analyzed, obtaining more data, informative and efficient, which would not be possible with only data cross section or time series. The results pointed to the existence of convergence-o-, ß- and ß--absolute-conditional, obtaining, in the case of absolute convergence, growth rates well above those found in the empirical literature. Under the use of proxies of human capital, the conditional convergence proved apparently differentiated between samples considered, revealing the persistence of differences interstate. / O presente trabalho analisa a hipotese da convergencia do PIB per capita nos estados nordestinos, no periodo de 1995-2009, com foco no capital humano. A necessidade de analisar apenas a Regiao Nordeste e nao todas as Regioes do Brasil prendeu-se a satisfazer a condicao de homogeneidade, necessaria para tratar a questao da convergencia. Tendo como aparato as teorias e modelos de crescimento economico, foram testadas as hipoteses de convergencia ß (absoluta e condicional) e convergencia-o- durante o periodo analisado. Utilizando o Software Stata versao 12.1, os coeficientes foram estimados atraves de dados em painel com efeitos fixos, que tem como principal vantagem captar as mudancas no comportamento dos indicadores analisados, obtendo dados mais, informativos e eficientes, o que nao seria possivel apenas com dados cross section ou com series temporais. Os resultados apontaram para a existencia de convergencia-o-, ß-absoluta e ß-condicional, obtendo, no caso da convergencia absoluta, taxas de crescimento bem superiores as encontradas na literatura empirica. Sob a utilizacao de proxies do capital humano, a convergencia condicional mostrou-se aparentemente diferenciada entre as amostras consideradas, revelando a persistencia das diferencas interestaduais.

There has been a continuous increase in the popularity of interactive gambling in South Africa due to the wide range of channels in which the public can now quickly and easily access gambling opportunities. Although there is uncertainty related to whether or not interactive gambling is considered to be legal in South Africa, the total value of income earned from interactive gambling by South African residents is conservatively estimated to be between R4 billion and R10 billion per year. Unfortunately, due to the fact that interactive gambling does not have national boundaries, it is difficult to determine the true worth of the industry. There appears to be no academic research which has examined whether income received from interactive gambling qualifies to be taxed in terms of the Income Tax Act 58 of 1962. This study will contribute to an academic understanding of the interactive gambling industry in South Africa and an understanding of the South African Revenue Service’s (SARS’s) policies and practices relating to the taxation of income earned from interactive gambling by resident South African individuals. The purpose of this study was to critically analyse whether income earned from interactive gambling by resident South African individuals should be taxed by SARS. This study also investigated which laws currently regulate the interactive gambling industry within South Africa and how the regulation and taxation of interactive gambling in South Africa compares to that of other developing and developed countries. This study also investigated the legality and estimated value of interactive gambling income earned in South Africa by resident South African individuals. This study concluded that income earned from interactive gambling should be subject to either normal income tax or to capital gains tax (CGT), depending on the intention of the taxpayer. This study noted that enforcing regular reporting from local licensed interactive gambling service providers would be the best method of effectively collecting tax that is owed by interactive gamblers on income generated from this form of gambling in South Africa. This study further concluded that the most appropriate regulatory model to be applied in South Africa would be a protectionist model. This model protects the residents of a country by having a regulated interactive gambling industry that is protected from outside intruders. A regulated industry will result in economic benefits such as taxation, investment and employment within South Africa. The difficulties associated with electronic-commerce could be overcome through regulation of the industry which would be beneficial to SARS. AFRIKAANS : Inter-aktiewe dobbel onder die Suid-Afrikaanse publiek is baie populêr en het ‘n voortgesette toename vanweë die beskikbaarheid en toeganklikheid wat deur tegnologie daargestel word. Alhoewel daar onsekerheid is aangaande of inter-aktiewe dobbel wettig is al dan nie, word die inkomste wat deur Suid Afrikaanse inwoners verdien word konserwatief geraam en beloop tussen R4 en R10 biljoen per jaar. Aangesien die toeganklikheid wat deur die internet gebied word nie net tot Suid-Afrika beperk is nie, maar wêreldwyd strek, is dit feitlik onmoonlik om die waarde van die industrie te bepaal. Daar is tot op hede geen akademiese navorsing gedoen om te bepaal of inkomste verdien uit inter-aktiewe dobbel belasbaar is ingevolge die Inkomstebelastingwet 58 van 1962. Hierdie studie sal bydra tot ‘n akademiese begrip van SARS se beleid en die toepassing daarvan op inkomste verdien uit inter-aktiewe dobbel deur Suid-Afrikaanse individue. Dus was die doel van hierdie navorsingstudie om te bepaal of inkomste verdien uit interaktiewe dobbel deur ‘n Suid Afrikaanse individu deur SARS belasbaar is of nie en die aspekte krities te analiseer. Hierdie studie ondersoek ook die wette wat huidiglik die belasbaarheid van inkomste uit inter-aktiewe dobbel in Suid Afrika reguleer asook hoe die regulasies vergelyk met die van ander ontwikkelende en onwikkelde lande. Verder word ondersoek ingestel om die wettigheid van inter-aktiewe dobbel en die waarde daarvan te bepaal. Die gevolgtrekking is dat inkomste verdien uit inter-aktiewe dobbel onderhewig moet wees aan of normale belasting of kapitaalwinsbelasting. Die navorsing en studie toon ook dat daar gelisensieerde inter-aktiewe dobbel diensverskaffers moet wees wat dan die invordering van die inkomste uit hierdie tipe dobbel kan beheer en dat inter-aktiewe dobbelaars wel die nodige belasting oorbetaal uit inkomste verdien in Suid-Afrika op hierdie manier. Ter afsluiting van die navorsing word daar ‘n model voorgestel wat toegepas moet word in Suid-Afrika, eerstens om die inwoners van die land te beskerm deur ‘n goed-gereguleerde inter-aktiewe dobbel stelsel daar te stel en tweedens moet daar ook die nodige beskerming verleen word teen buitelandse indringers. ‘n Goed gereguleerde model vir die industrie sal voordelig wees vir Suid Afrika deurdat die belasbaarheid van die inter-aktiewe dobbelinkomste ‘n inspuiting vir die ekonomie kan wees, sowel as werkskeppingvoordele bied. Wetstoepassing en regulering van die industrie sal ook voordelig wees vir SARS aangesien die probleme wat nou geassosieer word met elektroniese-handel voorkom en tot die minimum beperk kan word. / Dissertation (MCom)--University of Pretoria, 2010. / Taxation / unrestricted

Algunas especies vegetales se consideran malezas debido a que alcanzan niveles poblacionales tales, que al competir con los cultivos causan perjuicios sobre su producción. Por este motivo, en determinadas ocasiones es necesario realizar operaciones de control sobre las poblaciones de malezas para reducir su infestación hasta valores que permitan obtener niveles de producción deseados. A nivel mundial el control químico ha sido el método más práctico y eficiente para controlar malezas, especialmente en cultivos extensivos. Sin embargo, el uso continuo de herbicidas desde hace más de 40 años trajo aparejado un impacto negativo sobre el medio ambiente, la salud y la proliferación de malezas resistentes. Estas consecuencias se deben tener en cuenta para definir manejos adecuados a largo plazo. Desde una perspectiva estratégica, un programa sostenible de manejo de malezas debería basarse en una combinación de métodos tanto preventivos como curativos que apliquen principios basados en el conocimiento, dando lugar a lo que se conoce como Manejo Integrado de Malezas, MIM. En este contexto, el modelado matemático se presenta como una herramienta apropiada para ayudar a guiar el proceso de toma de decisiones asociado al MIM. En esta tesis se propone un modelo de simulación para asistir en la toma de decisiones relacionadas con el MIM. El modelo se desarrolló en colaboración con profesionales extensionistas y fue concebido como una herramienta flexible y adaptable a diversos sistemas de producción agrícola, poseyendo un mayor nivel de detalle que otros modelos similares. Dicho modelo permite simular la dinámica multianual de una maleza en competencia con el cultivo. El ciclo de vida de la maleza se representa a través de los componentes demográficos típicos (ej. banco de semillas, plántulas, adultos en estado vegetativo y reproductivo, producción de semillas). El desarrollo del cultivo se simula de forma simplificada a fin de cuantificar diariamente los efectos de la competencia interespecífica. La simulación de distintas estrategias de MIM permite calcular y comparar indicadores económicos, ambientales y agronómicos. Palabras claves: Manejo Integrado de Malezas, Modelado de sistemas agrícolas, Avena fatua L., Euphorbia davidii Subils., Trigo, Cebada, Soja, Competencia de malezas, Asistencia a la toma de decisiones, Margen bruto, Valor actual, Impacto ambiental, Malezas resistentes, Dinámica poblacional de malezas. / Some species considered weeds reach population levels that impact on crops’ production. For this reason, on certain occasions, it is necessary to carry out control operations to reduce weed infestations to levels compatible with reasonable crops’ yields. Worldwide, chemical control has been the most practical and efficient method for controlling weeds, especially in extensive crops. However, the continuous use of herbicides for over 40 years generated a well-known negative impact on environment and health and the proliferation of resistant species. These environmental consequences must be taken into account to define appropriate long-term management practices. From a strategic perspective, a sustainable weed management program should be based on a combination of preventive and curative methods that apply knowledge-based principles, leading to as the so-called Integrated Weed Management (IWM). In this context, mathematical modelling arises as an appropriate tool to assist in the decision- making process associated with IWM. This thesis proposes a simulation model to support decision-making related to IWM. The model was developed in collaboration with extensionists and was conceived as a flexible and adaptable tool for various agricultural production systems, possessing a higher level of detail than similar models. This model simulates multi-year dynamics of a weed in competition with a crop. The weed's life cycle is represented through typical demographic components (e.g. seed bank, seedlings, vegetative and reproductive individuals, and seed production). The development of the crop is simulated in a simplified manner in order to quantify the daily effects of interspecific competition. The simulation of different IWM strategies allows for the quantification and comparison of economic, environmental, and agronomic indexes.

The purpose of this work was to present and apply an analytical tool to the flows of goods and income between economic agents and the environment that allows us to calculate the Environmental Gross Domestic Product (EGDP) for Brazilian Legal Amazon. In order to achieve this goal the Environmental Social Accounting Matrix (ESAM) was developed. The model was developed from the traditional Social Accounting Matrix, the Environmental Input-Output Analysis Models, and from the United Nations handbook on the System of Integrated Environmental and Economic Accountings. The EGDP was calculated from the Environmental Input-Output Analysis, since the unavailability of data did not allow the application of the model of ESAM. The flows between the economy and the environment considered were the emissions of green house gases (depreciation of natural capital) and the investment needed to return the air to the same quality it had before being polluted. The results showed that the inclusion of depreciation of natural capital in the Gross Domestic Product (GDP) and in the added value (AV) calculation for each industry of the Brazilian Legal Amazon Region diminished the AV significantly in the industries of cattle (235%), soybean (77%), and other activities of livestock and agricultural (24%). In the Rest of Brazil, the industries with the highest impact of depreciation of natural capital on AV were soybean (30%), cattle (15%) and sugarcane (13%). The total EGDP of the Brazilian Legal Amazon Region was 15% smaller than its total GDP, whereas this difference for the rest of Brazil was 1%. Considering that the GDP is a component of economic wellbeing, the results show a significant reduction in economic wellbeing due to green house gas emissions released into the air by economic activities, mainly due to land use changes. The Environmental Social Accounting Matrix Model is a useful tool to help decision makers since it offers an analytical instrument for economic behavior and the impacts of economic activities on the environment. / The purpose of this work was to present and apply an analytical tool to the flows of goods and income between economic agents and the environment that allows us to calculate the Environmental Gross Domestic Product (EGDP) for Brazilian Legal Amazon. In order to achieve this goal the Environmental Social Accounting Matrix (ESAM) was developed. The model was developed from the traditional Social Accounting Matrix, the Environmental Input-Output Analysis Models, and from the United Nations handbook on the System of Integrated Environmental and Economic Accountings. The EGDP was calculated from the Environmental Input-Output Analysis, since the unavailability of data did not allow the application of the model of ESAM. The flows between the economy and the environment considered were the emissions of green house gases (depreciation of natural capital) and the investment needed to return the air to the same quality it had before being polluted. The results showed that the inclusion of depreciation of natural capital in the Gross Domestic Product (GDP) and in the added value (AV) calculation for each industry of the Brazilian Legal Amazon Region diminished the AV significantly in the industries of cattle (235%), soybean (77%), and other activities of livestock and agricultural (24%). In the Rest of Brazil, the industries with the highest impact of depreciation of natural capital on AV were soybean (30%), cattle (15%) and sugarcane (13%). The total EGDP of the Brazilian Legal Amazon Region was 15% smaller than its total GDP, whereas this difference for the rest of Brazil was 1%. Considering that the GDP is a component of economic wellbeing, the results show a significant reduction in economic wellbeing due to green house gas emissions released into the air by economic activities, mainly due to land use changes. The Environmental Social Accounting Matrix Model is a useful tool to help decision makers since it offers an analytical instrument for economic behavior and the impacts of economic activities on the environment.

Made available in DSpace on 2014-12-17T15:42:01Z (GMT). No. of bitstreams: 1 IedaNSM.pdf: 5948604 bytes, checksum: dbf245a349dd04cff535e8ba2e89601d (MD5) Previous issue date: 2007-07-06 / The aim of this work was to study a series of 11 different compositions of Ti-Zr binary alloys resistance to aggressive environment, i. e., their ability to keep their surface properties and mass when exposed to them as a way to evaluate their performance as biomaterials. The first stage was devoted to the fabrication of tablets from these alloys by Plasma-Skull casting method using a Discovery Plasma machine from EDG Equipamentos, Brazil. In a second stage, the chemical composition of each produced tablet was verified. In a third stage, the specimen were submitted to: as-cast microstructure analysis via optical and scanning electron microscopy (OM and SEM), x-ray dispersive system (EDS) chemical analysis via SEM, Vickers hardness tests for mechanical evaluation and corrosion resistence tests in a 0.9% NaCl solution to simulate exposition to human saliva monitored by open circuit potential and polarization curves. From the obtained results, it was possible to infer that specimens A1 (94,07 wt% Ti and 5,93% wt% Zr), A4 (77,81 wt % Ti and 22,19 wt % Zr) and A8 (27,83 wt% Ti and 72,17 wt% Zr), presented best performance regarding to corrosion resistance, homogeneity and hardness which are necessary issues for biomaterials to be applied as orthopedic and odontological prosthesis / Este trabalho teve como finalidade estudar uma s?rie de onze ligas met?licas bin?rias, tit?nio zirc?nio, quanto ? capacidade de resistirem ?s deteriora??es sem sofrerem modifica??es de suas propriedades iniciais de massa e superf?cie quando expostas a meios agressivos, como condi??o fundamental para que as mesmas sejam aplicadas como biomateriais. A primeira etapa tratou da confec??o das amostras das ligas em forma de pastilhas, empregando-se o processo de fundi??o Plasma Skull realizado na m?quina Discovery Plasma, produzida no Brasil pela EDG Equipamentos. Na segunda etapa, as amostras confeccionadas foram submetidas aos ensaios para a determina??o da composi??o qu?mica resultante de cada liga da s?rie ap?s a fus?o. E durante a terceira etapa foram feitos os procedimentos de avalia??es das amostras quanto ? homogeneidade e apresenta??o da microestrutura de cada uma das ligas no estado bruto de fus?o, como resultados dos ensaios metalogr?ficos por microscopia ?ptica e microscopia eletr?nica de varredura acoplada a espectroscopia por energia dispersiva de raios-x; quanto ? propriedade mec?nica de dureza medida em escala Vickers (HV); e quanto ?s propriedades qu?micas de resist?ncia ? corros?o, quando expostas em solu??o de cloreto de s?dio 0,9 %, para simular a saliva, empregando-se o monitoramento do potencial de circuito aberto e as curvas de polariza??o. A partir dos resultados obtidos foi poss?vel identificar as pastilhas de composi??es qu?micas A1 (94,07% de Ti e 5,93% de Zr), A4 (77,81% de Ti e 22,19% de Zr) e A8 (27,83% de Ti e 72,17% de Zr) que resultaram como maiores detentoras do conjunto de propriedades de resist?ncia ? corros?o, homogeneidade e dureza, as quais s?o mais necess?rias em biomateriais tipo pr?teses ortop?dicas ou odontol?gicas

Esta tese apresenta três ensaios sobre previsão de inflação e análise de dados em tempo real no Brasil. Utilizando uma curva de Phillips, o primeiro ensaio propõe um “modelo evolucionário” para prever inflação no Brasil. O modelo evolucionário consiste em uma combinação de um modelo não-linear (que é formado pela combinação de três redes neurais artificiais – RNAs) e de um modelo linear (que também é a referência para propósitos de comparação). Alguns parâmetros do modelo evolucionário, incluindo os pesos das combinações, evoluem ao longo do tempo segundo ajustes definidos por três algoritmos que avaliam os erros fora-da-amostra. As RNAs foram estimadas através de uma abordagem híbrida baseada em um algoritmo genético (AG) e em um algoritmo simplex de Nelder-Mead. Em um experimento de previsão fora-da-amostra para 3, 6, 9 e 12 passos à frente, o desempenho do modelo evolucionário foi comparado ao do modelo linear de referência, segundo os critérios de raiz do erro quadrático médio (REQM) e de erro absoluto médio (EAM). O desempenho do modelo evolucionário foi superior ao desempenho do modelo linear para todos os passos de previsão analisados, segundo ambos os critérios. O segundo ensaio é motivado pela recente literatura sobre análise de dados em tempo real, que tem mostrado que diversas medidas de atividade econômica passam por importantes revisões de dados ao longo do tempo, implicando importantes limitações para o uso dessas medidas. Elaboramos um conjunto de dados de PIB em tempo real para o Brasil e avaliamos a extensão na qual as séries de crescimento do PIB e de hiato do produto são revisadas ao longo do tempo. Mostramos que as revisões de crescimento do PIB (trimestre/trimestre anterior) são economicamente relevantes, embora as revisões de crescimento do PIB percam parte da importância à medida que o período de agregação aumenta (por exemplo, crescimento em quatro trimestres). Para analisar as revisões do hiato do produto, utilizamos quatro métodos de extração de tendência: o filtro de Hodrick-Prescott, a tendência linear, a tendência quadrática, e o modelo de Harvey-Clark de componentes não-observáveis. Todos os métodos apresentaram revisões de magnitudes economicamente relevantes. Em geral, tanto a revisão de dados do PIB como a baixa precisão das estimativas de final-de-amostra da tendência do produto mostraram-se fontes relevantes das revisões de hiato do produto. O terceiro ensaio é também um estudo de dados em tempo real, mas que analisa os dados de produção industrial (PI) e as estimativas de hiato da produção industrial. Mostramos que as revisões de crescimento da PI (mês/mês anterior) e da média móvel trimestral são economicamente relevantes, embora as revisões de crescimento da PI tornem-se menos importantes à medida que o período de agregação aumenta (por exemplo, crescimento em doze meses). Para analisar as revisões do hiato da PI, utilizamos três métodos de extração de tendência: o filtro de Hodrick-Prescott, a tendência linear e a tendência quadrática. Todos os métodos apresentaram revisões de magnitudes economicamente relevantes. Em geral, tanto a revisão de dados da PI como a baixa precisão das estimativas de final-de-amostra da tendência da PI mostraram-se fontes relevantes das revisões de hiato da PI, embora os resultados sugiram certa predominância das revisões provenientes da baixa precisão de final-de-amostra. / This thesis presents three essays on inflation forecasting and real-time data analysis in Brazil. By using a Phillips curve, the first essay presents an “evolutionary model” to forecast Brazilian inflation. The evolutionary model consists in a combination of a non-linear model (that is formed by a combination of three artificial neural networks - ANNs) and a linear model (that is also a benchmark for comparison purposes). Some parameters of the evolutionary model, including the combination weight, evolve throughout time according to adjustments defined by three algorithms that evaluate the out-of-sample errors. The ANNs were estimated by using a hybrid approach based on a genetic algorithm (GA) and on a Nelder-Mead simplex algorithm. In a 3, 6, 9 and 12 steps ahead out-of-sample forecasting experiment, the performance of the evolutionary model was compared to the performance of the benchmark linear model, according to root mean squared errors (RMSE) and to mean absolute error (MAE) criteria. The evolutionary model performed better than the linear model for all forecasting steps that were analyzed, according to both criteria. The second essay is motivated by recent literature on real-time data analysis, which has shown that several measures of economic activities go through important data revisions throughout time, implying important limitations to the use of these measures. We developed a GDP real-time data set to Brazilian economy and we analyzed the extent to which GDP growth and output gap series are revised over time. We showed that revisions to GDP growth (quarter-onquarter) are economic relevant, although the GDP growth revisions lose part of their importance as aggregation period increases (for example, four-quarter growth). To analyze the output gap revisions, we applied four detrending methods: the Hodrick-Prescott filter, the linear trend, the quadratic trend, and the Harvey-Clark model of unobservable components. It was shown that all methods had economically relevant magnitude of revisions. In a general way, both GDP data revisions and the low accuracy of end-of-sample output trend estimates were relevant sources of output gap revisions. The third essay is also a study about real-time data, but focused on industrial production (IP) data and on industrial production gap estimates. We showed that revisions to IP growth (month-on-month) and to IP quarterly moving average growth are economic relevant, although the IP growth revisions become less important as aggregation period increases (for example, twelve-month growth). To analyze the output gap revisions, we applied three detrending methods: the Hodrick-Prescott filter, the linear trend, and the quadratic trend. It was shown that all methods had economically relevant magnitude of revisions. In general, both IP data revisions and low accuracy of end-of-sample IP trend estimates were relevant sources of IP gap revisions, although the results suggest some prevalence of revisions originated from low accuracy of end-of-sample estimates.

Submitted by Cássio da Cruz Nogueira (cassionogueirakk@gmail.com) on 2017-02-08T12:43:22Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertacao_ImpactosDesastresNaturais.pdf: 1989991 bytes, checksum: abbc4fb8f3cbfdcf343efccc6b82a1db (MD5) / Approved for entry into archive by Edisangela Bastos (edisangela@ufpa.br) on 2017-02-10T17:56:38Z (GMT) No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertacao_ImpactosDesastresNaturais.pdf: 1989991 bytes, checksum: abbc4fb8f3cbfdcf343efccc6b82a1db (MD5) / Made available in DSpace on 2017-02-10T17:56:38Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Dissertacao_ImpactosDesastresNaturais.pdf: 1989991 bytes, checksum: abbc4fb8f3cbfdcf343efccc6b82a1db (MD5) Previous issue date: 2016 / Nos últimos anos, tem aumentado a frequência dos desastres naturais na Amazônia, concomitantemente com a necessidade de estimar os prejuízos econômicos destes eventos, no intuito de definir estratégias para o desenvolvimento da região. Este trabalho tem como objetivo verificar qual é o tipo e intensidade do impacto das inundações de 2009, consideradas as maiores já registradas na Amazônia, ao crescimento do Produto Interno Bruto (PIB) municipal. A metodologia consistiu na análise de correlação linear entre o desempenho econômico (calculado com base no PIB observado e esperado para os anos de 2009 a 2012) e dois conjuntos de variáveis: qualitativas (municípios atingidos por inundação, tipo de inundação, estado de localização) e quantitativas (PIB, tamanho geográfico, população, PIB per capita, densidade populacional, Índice de Desenvolvimento Humano Municipal, renda média, índices de pobreza humana e Gini, bem como as receitas e transferências orçamentárias para os municípios). Os resultados mostram que as inundações bruscas são as que mais afetam o crescimento econômico de longo prazo, constituindo um potencial problema para o desenvolvimento. As principais estratégias a serem adotadas perpassam pela inclusão de ações de mitigação de desastres naturais ao planejamento do crescimento econômico e ao desenvolvimento sustentável da Amazônia. / In the last years, have been increase the frequency of natural disasters in Amazon, concomitantly whit the need of estimate the economic losses these events, in order of define strategies to regional development. This work have the objective of check what is the type and intensity of 2009 floods impacts, considered the major already registered in Amazon, to the increase of municipal Gross Domestic Product (GDP). The methodology was the linear correlation analysis between the economic performance (calculated whit base in observed and expected GDP to years of 2009 to 2012) and two sets of variables: qualitative (municipalities hit by floods, type of flood, state of localization) and quantitative (GDP, geographic size, population, GDP per capita, population density, Indices of Municipal Human Development, average income, indices of human poor and Gini, as well as income and budget transfers to municipalities). The results bring which the fast flood are the which more affect the long term economic increase, constituting a potential problem to development. The means strategies to be adopted are the inclusion of natural disasters mitigation actions in plan of economics increase and to sustainable development of Amazon. / En los últimos años, tiene aumentado la frecuencia de los desastres naturales en la Amazonia, concomitantemente con la necesidad de estimar los prejuicios económicos distes eventos, en orden de definir estrategias para el desenvolvimiento de la región. Este trabajo tiene como objetivo verificar cuál es el tipo e intensidad do impacto de las inundaciones de 2009, consideradas las mayores ya registradas en la Amazonia, a el crecimiento de lo Producto Interno Bruto (PIB) municipal. La metodología consistió na análisis de correlación linear entre el desempeño económico (calculado con base en el PIB observado y esperado para los años de 2009 hasta 2012) y dos conjuntos de variables: cualitativas (municipios atingidos por inundaciones, tipo de inundación, estado de localización) e cuantitativas (PIB, tamaño geográfico, populación, PIB per capita, densidad de población, Índice de Desenvolvimiento Humano Municipal, renda media, índices de pobreza y Gini, así como las transferencias de ingresos y de presupuesto a los municipios). Los resultados muestran que las inundaciones bruscas son las que más afectan el crecimiento económico de lo curto plazo, constituyendo un potencial problema para el desenvolvimiento. Las principales estrategias para ser adoptadas son la inclusión de acciones de mitigación de desastres naturales al planeamiento de lo crecimiento económico e al desenvolvimiento sustentable de la Amazonia. / Ces dernières années, a augmenté la fréquence des catastrophes naturelles en Amazonie, en même temps que la nécessité d'estimer les pertes économiques de ces événements, afin d'élaborer des stratégies pour le développement de la région. Ce travail vise à vérifier quelle est la nature et l'intensité de l'impact des inondations de 2009, considéré comme le plus important jamais enregistré dans l'Amazonie, la croissance du Produit Intérieur Brut (PIB). La méthodologie a consisté à l'analyse de corrélation linéaire entre la performance économique (calculé sur la base du PIB observé et attendu pour les années 2009 à 2012) et deux ensembles de variables : qualitative (municipalités touchées par les inondations, type de les inondations, province de localization) et quantitative (PIB, superficie, population, PIB par habitant, la densité de population, l'indicateur du développement humain Municipal, revenu moyen par habitant les indices de pauvreté humaine et Gini, ainsi que le revenu et budgétaires transferts aux municipalités). Les résultats montrent que les inondations soudaines sont celles qui affectent la croissance économique à long terme, qui constitue un problème potentiel pour le développement. Les principales stratégies arrêtées en pâturage, par l'inclusion des catastrophes naturelles, des mesures d'atténuation à la planification de la croissance économique et le développement durable de l'Amazonie.

Esta tese apresenta três ensaios sobre previsão de inflação e análise de dados em tempo real no Brasil. Utilizando uma curva de Phillips, o primeiro ensaio propõe um “modelo evolucionário” para prever inflação no Brasil. O modelo evolucionário consiste em uma combinação de um modelo não-linear (que é formado pela combinação de três redes neurais artificiais – RNAs) e de um modelo linear (que também é a referência para propósitos de comparação). Alguns parâmetros do modelo evolucionário, incluindo os pesos das combinações, evoluem ao longo do tempo segundo ajustes definidos por três algoritmos que avaliam os erros fora-da-amostra. As RNAs foram estimadas através de uma abordagem híbrida baseada em um algoritmo genético (AG) e em um algoritmo simplex de Nelder-Mead. Em um experimento de previsão fora-da-amostra para 3, 6, 9 e 12 passos à frente, o desempenho do modelo evolucionário foi comparado ao do modelo linear de referência, segundo os critérios de raiz do erro quadrático médio (REQM) e de erro absoluto médio (EAM). O desempenho do modelo evolucionário foi superior ao desempenho do modelo linear para todos os passos de previsão analisados, segundo ambos os critérios. O segundo ensaio é motivado pela recente literatura sobre análise de dados em tempo real, que tem mostrado que diversas medidas de atividade econômica passam por importantes revisões de dados ao longo do tempo, implicando importantes limitações para o uso dessas medidas. Elaboramos um conjunto de dados de PIB em tempo real para o Brasil e avaliamos a extensão na qual as séries de crescimento do PIB e de hiato do produto são revisadas ao longo do tempo. Mostramos que as revisões de crescimento do PIB (trimestre/trimestre anterior) são economicamente relevantes, embora as revisões de crescimento do PIB percam parte da importância à medida que o período de agregação aumenta (por exemplo, crescimento em quatro trimestres). Para analisar as revisões do hiato do produto, utilizamos quatro métodos de extração de tendência: o filtro de Hodrick-Prescott, a tendência linear, a tendência quadrática, e o modelo de Harvey-Clark de componentes não-observáveis. Todos os métodos apresentaram revisões de magnitudes economicamente relevantes. Em geral, tanto a revisão de dados do PIB como a baixa precisão das estimativas de final-de-amostra da tendência do produto mostraram-se fontes relevantes das revisões de hiato do produto. O terceiro ensaio é também um estudo de dados em tempo real, mas que analisa os dados de produção industrial (PI) e as estimativas de hiato da produção industrial. Mostramos que as revisões de crescimento da PI (mês/mês anterior) e da média móvel trimestral são economicamente relevantes, embora as revisões de crescimento da PI tornem-se menos importantes à medida que o período de agregação aumenta (por exemplo, crescimento em doze meses). Para analisar as revisões do hiato da PI, utilizamos três métodos de extração de tendência: o filtro de Hodrick-Prescott, a tendência linear e a tendência quadrática. Todos os métodos apresentaram revisões de magnitudes economicamente relevantes. Em geral, tanto a revisão de dados da PI como a baixa precisão das estimativas de final-de-amostra da tendência da PI mostraram-se fontes relevantes das revisões de hiato da PI, embora os resultados sugiram certa predominância das revisões provenientes da baixa precisão de final-de-amostra. / This thesis presents three essays on inflation forecasting and real-time data analysis in Brazil. By using a Phillips curve, the first essay presents an “evolutionary model” to forecast Brazilian inflation. The evolutionary model consists in a combination of a non-linear model (that is formed by a combination of three artificial neural networks - ANNs) and a linear model (that is also a benchmark for comparison purposes). Some parameters of the evolutionary model, including the combination weight, evolve throughout time according to adjustments defined by three algorithms that evaluate the out-of-sample errors. The ANNs were estimated by using a hybrid approach based on a genetic algorithm (GA) and on a Nelder-Mead simplex algorithm. In a 3, 6, 9 and 12 steps ahead out-of-sample forecasting experiment, the performance of the evolutionary model was compared to the performance of the benchmark linear model, according to root mean squared errors (RMSE) and to mean absolute error (MAE) criteria. The evolutionary model performed better than the linear model for all forecasting steps that were analyzed, according to both criteria. The second essay is motivated by recent literature on real-time data analysis, which has shown that several measures of economic activities go through important data revisions throughout time, implying important limitations to the use of these measures. We developed a GDP real-time data set to Brazilian economy and we analyzed the extent to which GDP growth and output gap series are revised over time. We showed that revisions to GDP growth (quarter-onquarter) are economic relevant, although the GDP growth revisions lose part of their importance as aggregation period increases (for example, four-quarter growth). To analyze the output gap revisions, we applied four detrending methods: the Hodrick-Prescott filter, the linear trend, the quadratic trend, and the Harvey-Clark model of unobservable components. It was shown that all methods had economically relevant magnitude of revisions. In a general way, both GDP data revisions and the low accuracy of end-of-sample output trend estimates were relevant sources of output gap revisions. The third essay is also a study about real-time data, but focused on industrial production (IP) data and on industrial production gap estimates. We showed that revisions to IP growth (month-on-month) and to IP quarterly moving average growth are economic relevant, although the IP growth revisions become less important as aggregation period increases (for example, twelve-month growth). To analyze the output gap revisions, we applied three detrending methods: the Hodrick-Prescott filter, the linear trend, and the quadratic trend. It was shown that all methods had economically relevant magnitude of revisions. In general, both IP data revisions and low accuracy of end-of-sample IP trend estimates were relevant sources of IP gap revisions, although the results suggest some prevalence of revisions originated from low accuracy of end-of-sample estimates.