Previsão do índice bovespa por meio de redes neurais artificiais: uma análise comparada aos métodos tradicionais de séries de tempo

Forecast is the basis for making strategic, tactical and operational business decisions. In financial economics, several techniques have been used to predict the behavior of assets over the past decades.Thus, there are several methods to assist in the task of time series forecasting, however, conventional modeling techniques such as statistical models and those based on theoretical mathematical models have produced unsatisfactory predictions, increasing the number of studies in more advanced methods of prediction. Among these, the Artificial Neural Networks (ANN) are a relatively new and promising method for predicting business that shows a technique that has caused much interest in the financial environment and has been used successfully in a wide variety of financial modeling systems applications, in many cases proving its superiority over the statistical models ARIMA-GARCH. In this context, this study aimed to examine whether the ANNs are a more appropriate method for predicting the behavior of Indices in Capital Markets than the traditional methods of time series analysis. For this purpose we developed an quantitative study, from financial economic indices, and developed two models of RNA-type feedfoward supervised learning, whose structures consisted of 20 data in the input layer, 90 neurons in one hidden layer and one given as the output layer (Ibovespa). These models used backpropagation, an input activation function based on the tangent sigmoid and a linear output function. Since the aim of analyzing the adherence of the Method of Artificial Neural Networks to carry out predictions of the Ibovespa, we chose to perform this analysis by comparing results between this and Time Series Predictive Model GARCH, developing a GARCH model (1.1).Once applied both methods (ANN and GARCH) we conducted the results' analysis by comparing the results of the forecast with the historical data and by studying the forecast errors by the MSE, RMSE, MAE, Standard Deviation, the Theil's U and forecasting encompassing tests. It was found that the models developed by means of ANNs had lower MSE, RMSE and MAE than the GARCH (1,1) model and Theil U test indicated that the three models have smaller errors than those of a naïve forecast. Although the ANN based on returns have lower precision indicator values than those of ANN based on prices, the forecast encompassing test rejected the hypothesis that this model is better than that, indicating that the ANN models have a similar level of accuracy . It was concluded that for the data series studied the ANN models show a more appropriate Ibovespa forecasting than the traditional models of time series, represented by the GARCH model

Navegação de robô móvel em ambiente com humanos

Trabalho Final de Mestrado para obtenção do grau de Mestre em Engenharia Mecânica

A NAIRU assessment: an applied case to Colombia’s labor markets

Given the discrepancy over the optimum levels of employment for Colombia, this research targets both, the national and urban, Non-Accelerating Inflation Rate of Unemployment (NAIRU) for the Colombian markets -- In doing so, there is a strong pertinence in estimating the constant NAIRU through raw and minimally altered data and providing the reader with a complete brief of the theory in which the model is founded -- The introduction of supply shocks is considered to attain improved estimations and a more reliable assessment of the NAIRU to those that have previously been attempted -- The backbone of the analysis is conducted through the relationship established by the Phillips curve from 2001 until 2015

Organizaciones virtuales y redes neuronales. Algunas similitudes.

Aunque existen numerosos trabajos que identifican sus principales características y modo de funcionamiento, el estudio de las organizaciones virtuales adolece de una carencia de modelos matemáticos que reflejen su comportamiento de un modo cuantitativo. En este sentido, a lo largo del presente trabajo se tratará de poner de manifiesto las similitudes existentes entre el funcionamiento de las organizaciones virtuales y el de las redes neuronales (SOM, SelfOrganizing Maps). El objetivo es sentar las bases para proponer este tipo de técnica estadística como herramienta para la formulación de modelos sobre organizaciones virtuales. Se plantearán una serie de argumentos de plausibilidad, dejando a investigaciones posteriores la verificación rigurosa de esta propuesta.

A generalized model for population dynamics where interactions change over time

La dinámica demográfica ha sido modelada con ecuaciones diferenciales desde que Malthus comenzó sus estudios hace más de doscientos años atrás. Los modelos convencionales siempre tratan relaciones entre especies como estáticas, denotando sólo su dependencia durante un período fijo del tiempo, aunque sea conocido que las relaciones entre especies pueden cambiar con el tiempo. Aquí proponemos un modelo para la dinámica demográfica que incorpora la evolución con el tiempo de las interacciones entre especies. Este modelo incluye una amplia gama de interacciones, de depredador-presa a las relaciones mutualistas, ya sea obligada o facultativa. El mecanismo que describimos permite la transición de una clase de relación entre especies a algún otro, según algunos parámetros externos fijados por el contexto. Estas transiciones podrían evitar la extinción de una de las especies, si esto termina por depender demasiado del ambiente o su relación con las otras especies.

Políticas para la integración del juicio experto y los pronósticos estadísticos en el marco organizacional

Ha sido ampliamente reconocida la importancia de la predicción en la toma de decisiones, y se han encontrado evidencias de que uno de los métodos más efectivos es el ajuste de los pronósticos obtenidos a partir de modelos matemáticos usando juicios informados. No obstante, existe una amplia cantidad de factores que pueden afectar la calidad y credibilidad de las predicciones; en este trabajo se examinan aquellos factores relacionados con las políticas organizacionales, y se proponen varias estrategias para su mitigación.

Análise da fluência em vigas de concreto armado simplesmente apoiadas usando o método dos elementos finitos

Reinforced concrete creep is a phenomenon of great importance. Despite being appointed as the main cause of several pathologies, its effects are yet considered in a simplified way by the structural designers. In addition to studying the phenomenon in reinforced concrete structures and its current account used in the structural analysis, this paper compares creep strains at simply supported reinforced concrete beams in analytical and in experimental forms with the finite element method (FEM) simulation results. The strains and deflections obtained through the analytical form were calculated with the Brazilian code NBR 6118 (2014) recommendations and the simplified method from CEB-FIP 90 and the experimental results were extracted from tests available in the literature. Finite element simulations are performed using ANSYS Workbench software, using its 3D SOLID 186 elements and the structure symmetry. Analyzes of convergence using 2D PLANE 183 elements are held as well. At the end, it is concluded that FEM analyses are quantitative and qualitative efficient for the estimation of this non-linearity and that the method utilized to obtain the creep coefficients values is sufficiently accurate.

Ocho diagnósticos para el desarrollo regional

El Programa de Intercambio Educativo (PIE) se inicia en el año 1987 como una red orientada a desarrollar vínculos institucionales entre la Universidad del Pacífico y universidades públicas de reconocida trayectoria y calidad académica, ubicadas en diversas regiones del país. Desde el inicio, el objetivo principal de la red PIE fue contribuir con la mejora de la calidad de la enseñanza y de la investigación de las instituciones que la conformaban, con un énfasis particular en las especialidades de Administración, Economía y Contabilidad. En tal sentido, el programa se fundamentó, desde su creación, en una filosofía orientada a promover la “formación de formadores”. Con el paso de los años, la red se fue abriendo a otras universidades del país y actualmente el nombre “PIE” ya no se refiere a una pequeña red de universidades, sino al programa de capacitación docente que ofrece la Universidad del Pacífico, con el apoyo de su Patronato y de sus empresas benefactoras, a los profesores de todas las universidades públicas del país. La presente publicación recoge las investigaciones realizadas por docentes –y en algunos casos estudiantes de último año– de Economía de diversas universidades del país, las cuales fueron seleccionadas para ser presentadas en la I y la II Conferencia Académica del Programa de Intercambio Educativo (PIE), realizadas en la Universidad del Pacífico los años 2012 y 2013, respectivamente.

Valoração do efluente da indústria processadora de pescado por incorporação de nutrientes em aphanothece microscopica nägeli

Nesta tese procurou-se demonstrar a valoração do efluente do processamento de pescado por incorporação dos nutrientes em Aphanothece microscopica Nägeli a diferentes temperaturas. Para tanto o trabalho é composto de cinco artigos que objetivaram avaliar sob o ponto de vista do tratamento do efluente pela cianobactéria Aphanothece e a separação e avaliação da biomassa gerada. O primeiro artigo intitula-se “Influência da temperatura na remoção de nutrientes do efluente da indústria de pescado por Aphanothece microscopica Nägeli”, e teve por objetivo avaliar a influência da temperatura (10, 20 e 30ºC) em um sistema de tratamento pela cianobactéria Aphanothece na remoção de matéria orgânica, nitrogênio e fósforo do efluente oriundo do processamento de pescado. A análise dos resultados mostrou que a temperatura influenciou significativamente na remoção de DQO, NTK, N-NH4 + e P-PO4 -3 . Para os experimentos a 20 e 30ºC todos os limites estabelecidos para os parâmetros avaliados foram atingidos. O segundo artigo intitulado “Efeito de coagulantes no efluente da indústria da pesca visando à separação de biomassa quando tratado por cianobactéria” avaliou o efeito da concentração e pH de dois tipos de coagulantes, cloreto férrico (FeCl3) e sulfato de alumínio (Al2(SO4)3), na separação da biomassa da cianobactéria Aphanothece microscopica Nägeli cultivada em efluente da indústria da pesca, assim como a remoção de matéria orgânica e nutrientes do efluente. Os resultados indicaram que o coagulante FeCl3 foi mais eficaz na remoção de todos os parâmetros testados. No que concerne à separação da biomassa, com um número de seis lavagens foi removido cerca de 97,6% da concentração de FeCl3 adicionado inicialmente. O terceiro artigo com o título “Caracterização da biomassa de Aphanothece microscopica Nägeli gerada no efluente da indústria da pesca em diferentes temperaturas de cultivo” avaliou a composição química da biomassa da cianobactéria Aphanothece microscopica Nägeli quando desenvolvida em meio de cultivo padrão BG11 e no efluente do processamento de pescado. O quarto artigo teve como título “Influência do meio de cultivo e temperatura em compostos nitrogenados na cianobactéria Aphanothece microscopica Nägeli” objetivou avaliar o teor de compostos nitrogenados presentes na biomassa da cianobactéria Aphanothece microscopica Nägeli quando cultivada em meio padrão e no efluente da indústria da pesca nas diferentes fases de crescimento. Para o estudo da composição química e nitrogenados no efluente foram realizados experimentos nas temperaturas de 10, 20 e 30ºC. As concentrações de proteína, cinzas e pigmentos aumentaram com o aumento da temperatura. Por outro lado, foi observada uma redução do teor de lipídios e carboidratos com o aumento da temperatura. O íon amônio juntamente com os ácidos nucléicos representa uma importante fração do nitrogênio não protéico presente na biomassa da cianobactéria Aphanothece. Ficou demonstrada a influência do meio de cultivo na concentração de nitrogênio, bem como a determinação de proteína pelo método de Kjeldahl superestima a concentração protéica em cianobactérias. O quinto artigo intitulado “Produção de proteína unicelular a partir do efluente do processamento do pescado: modelagem preditiva e simulação” avaliou a produção de proteína unicelular através do cultivo da cianobactéria Aphanothece microscopica Nägeli no efluente da indústria da pesca. Os dados cinéticos de crescimento celular foram ajustados a quatro modelos matemáticos (Logístico, Gompertz, Gompertz Modificado e Baranyi). Os resultados demonstraram que o modelo Logístico foi considerado o mais adequado para descrever a formação de biomassa. A análise preditiva mostrou a possibilidade da obtenção de 1,66, 18,96 e 57,36 kg.m-3.d-1 de biomassa por volume do reator em 1000 h de processo contínuo, para as temperaturas de 10, 20 e 30ºC, respectivamente.

Modelo matemático de reconciliación de datos volumétricos en el control de balances y mediciones de hidrocarburos refinados para el sistema de transporte pozos – Galán L14” de la vicepresidencia de transporte de Ecopetrol S.A

Tesis (Maestria en Ingeniería).-- Universidad de La Salle. Facultad de Ingeniería. Maestría en Ingeniería, 2014

Aplicação da metodologia numérica “fast bounds crack” para uma estimativa eficiente da evolução do tamanho de trinca

A significant part of the life of a mechanical component occurs, the crack propagation stage in fatigue. Currently, it is had several mathematical models to describe the crack growth behavior. These models are classified into two categories in terms of stress range amplitude: constant and variable. In general, these propagation models are formulated as an initial value problem, and from this, the evolution curve of the crack is obtained by applying a numerical method. This dissertation presented the application of the methodology "Fast Bounds Crack" for the establishment of upper and lower bounds functions for model evolution of crack size. The performance of this methodology was evaluated by the relative deviation and computational times, in relation to approximate numerical solutions obtained by the Runge-Kutta method of 4th explicit order (RK4). Has been reached a maximum relative deviation of 5.92% and the computational time was, for examples solved, 130,000 times more higher than achieved by the method RK4. Was performed yet an Engineering application in order to obtain an approximate numerical solution, from the arithmetic mean of the upper and lower bounds obtained in the methodology applied in this work, when you don’t know the law of evolution. The maximum relative error found in this application was 2.08% which proves the efficiency of the methodology "Fast Bounds Crack".

Análise e modelagem do escoamento transitório de fluidos de perfuração

Drilling fluids present a thixotropic behavior and they usually gel when at rest. The sol-gel transition is fundamental to prevent the deposit of rock fragments, generated by drilling the well, over the drill bit during eventual stops. Under those conditions, high pressures are then required in order to break-up the gel when circulation is resumed. Moreover, very high pressures can damage the rock formation at the bottom of the well. Thus, a better understanding of thixotropy and the behavior of thixotropic materials becomes increasingly important for process control. The mechanisms that control thixotropy are not yet well defined and modeling is still a challenge. The objective of this work is to develop a mathematical model to study the pressure transmission in drilling fluids. This work presents a review of thixotropy and of different mathematical models found in the literature that are used to predict such characteristic. It also shows a review of transient flows of compressible fluids. The problem is modeled as the flow between the drillpipe and the annular region (space between the wall and the external part of the drillpipe). The equations that describe the problem (mass conservation, momentum balance, constitutive and state) are then discretized and numerically solved by using a computational algorithm in Fortran. The model is validated with experimental and numerical data obtained from the literature. Comparisons between experimental data obtained from Petrobras and calculated by three viscoplastic and one pseudoplastic models are conducted. The viscoplastic fluids, due to the yield stress, do not fully transmit the pressure to the outlet of the annular space. Sensibility analyses are then conducted in order to evaluate the thixotropic effect in pressure transmission.

Comparação do controle do inversor trifásico conectado à rede com filtro LCL considerando o amortecimento passivo e ativo

Distributed generation systems must fulfill standards specifications of current harmonics injected to the grid. In order to satisfy these grid requirements, passive filters are connected between inverter and grid. This work compares the characteristic response of the traditional inductive (L) filter with the inductive-capacitive-inductive (LCL) filter. It is shown that increasing the inductance L leads to a good ripple current suppression around the inverter switching frequency. The LCL filter provides better harmonic attenuation and reduces the filter size. The main drawback is the LCL filter impedance, which is characterized by a typical resonance peak, which must be damped to avoid instability. Passive or active techniques can be used to damp the LCL resonance. To address this issue, this dissertation presents a comparison of current control for PV grid-tied inverters with L filter and LCL filter and also discuss the use of active and passive damping for different regions of resonance frequency. From the mathematical models, a design methodology of the controllers was developed and the dynamic behavior of the system operating in closed loop was investigated. To validate the studies developed during this work, experimental results are presented using a three-phase 5kW experimental platform. The main components and their functions are discussed in this work. Experimental results are given to support the theoretical analysis and to illustrate the performance of grid-connected PV inverter system. It is shown that the resonant frequency of the system, and sampling frequency can be associated in order to calculate a critical frequency, below which is essential to perform the damping of the LCL filter. Also, the experimental results show that the active buffer per virtual resistor, although with a simple development, is effective to damp the resonance of the LCL filter and allow the system to operate stable within predetermined parameters.