Estimação paramétrica e não-paramétrica em modelos de markov ocultos

In this work we study the Hidden Markov Models with finite as well as general state space. In the finite case, the forward and backward algorithms are considered and the probability of a given observed sequence is computed. Next, we use the EM algorithm to estimate the model parameters. In the general case, the kernel estimators are used and to built a sequence of estimators that converge in L1-norm to the density function of the observable process

O uso de modelos ocultos de Markov no estudo do fluxo de rios intermitentes

In this work, we present our understanding about the article of Aksoy [1], which uses Markov chains to model the flow of intermittent rivers. Then, we executed an application of his model in order to generate data for intermittent streamflows, based on a data set of Brazilian streams. After that, we build a hidden Markov model as a proposed new approach to the problem of simulation of such flows. We used the Gamma distribution to simulate the increases and decreases in river flows, along with a two-state Markov chain. The motivation for us to use a hidden Markov model comes from the possibility of obtaining the same information that the Aksoy’s model provides, but using a single tool capable of treating the problem as a whole, and not through multiple independent processes

Crenças normativas e comportamentais e suas influências na intenção de adoção de um sistema de informação: um estudo na área de compras da UFPA

The study aims to identify the factors that influence the behavior intention to adopt an academic Information System (SIE), in an environment of mandatory use, applied in the procurement process at the Federal University of Pará (UFPA). For this, it was used a model of innovation adoption and technology acceptance (TAM), focused in attitudes and intentions regarding the behavior intention. The research was conducted a quantitative survey, through survey in a sample of 96 administrative staff of the researched institution. For data analysis, it was used structural equation modeling (SEM), using the partial least squares method (Partial Least Square PLS-PM). As to results, the constructs attitude and subjective norms were confirmed as strong predictors of behavioral intention in a pre-adoption stage. Despite the use of SIE is required, the perceived voluntariness also predicts the behavior intention. Regarding attitude, classical variables of TAM, like as ease of use and perceived usefulness, appear as the main influence of attitude towards the system. It is hoped that the results of this study may provide subsidies for more efficient management of the process of implementing systems and information technologies, particularly in public universities

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

Análise numérica de dutos enterrados sujeitos a elevação e recalque localizado

Problems associated to longitudinal interactions in buried pipelines are characterized as three-dimensional and can lead to different soil-pipe issues. Despite the progress achieved in research on buried pipelines, little attention has been given to the three-dimensional nature of the problem throughout the last decades. Most of researches simplify the problem by considering it in plane strain condition. This dissertation aims to present a study on the behavior of buried pipelines under local settlement or elevation, using three-dimensional simulations. Finite element code Plaxis 3D was used for the simulations. Particular aspects of the numerical modeling were evaluated and parametric analyzes were performed, was investigated the effects of soil arching in three-dimensional form. The main variables investigated were as follows: relative density, displacement of the elevation or settlement zone, elevated zone size, height of soil cover and pipe diameter/thickness ratio. The simulations were performed in two stages. The first stage was involved the validation of the numerical analysis using the physical models put forward by Costa (2005). In the second stage, numerical analyzes of a full-scale pipeline subjected to a localized elevation were performed. The obtained results allowed a detailed evaluation of the redistribution of stresses in the soil mass and the deflections along the pipe. It was observed the reduction of stresses in the soil mass and pipe deflections when the height of soil cover was decreased on regions of the pipe subjected to elevation. It was also shown for the analyzed situation that longitudinal thrusts were higher than vi circumferential trusts and exceeded the allowable stresses and deflections. Furthermore, the benefits of minimizing stress with technical as the false trench, compressible cradle and a combination of both applied to the simulated pipeline were verified

Crenças normativas e comportamentais e suas influências na intenção de adoção de um sistema de informação: um estudo na área de compras da UFPA

The study aims to identify the factors that influence the behavior intention to adopt an academic Information System (SIE), in an environment of mandatory use, applied in the procurement process at the Federal University of Pará (UFPA). For this, it was used a model of innovation adoption and technology acceptance (TAM), focused in attitudes and intentions regarding the behavior intention. The research was conducted a quantitative survey, through survey in a sample of 96 administrative staff of the researched institution. For data analysis, it was used structural equation modeling (SEM), using the partial least squares method (Partial Least Square PLS-PM). As to results, the constructs attitude and subjective norms were confirmed as strong predictors of behavioral intention in a pre-adoption stage. Despite the use of SIE is required, the perceived voluntariness also predicts the behavior intention. Regarding attitude, classical variables of TAM, like as ease of use and perceived usefulness, appear as the main influence of attitude towards the system. It is hoped that the results of this study may provide subsidies for more efficient management of the process of implementing systems and information technologies, particularly in public universities

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

Uma contribuição crítica para o redesenho de cadeiras de rodas adaptadas para crianças e adolescentes com paralisia cerebral

This work was aimed at making a critical analysis of the product wheelchair, both for using four different models, which were objects of study of the dissertation of Cláudia Regina Cabral Galvão, entitled Critical Analysis of the Mobility Products Seated a wheelchair - Used by Children and Adolescents with Cerebral Palsy in Natal / RN and other municipalities of Rio Grande do Norte . This product is considered an instrument in the social rehabilitation of great importance for people with physical disabilities. This study aims to position the issue and develop comments on technical up grading of certain models according to the needs of the user. Describes features of four models in search searched through catalogs in order to know its advantages and disadvantages of use. Were presented the definitions of ergonomics and ergonomic aspects to be considered on a design, the study of anthropometry and its recommendations. Discussions the methodology of project design in two parts: the first, on the structuring of design problem (formulation, analysis, synthesis and evaluation.) And the second on the project (design and development, implementation and evaluation and solution). With that review will include the possibilities for a new redesign of the wheelchair, based on forms of adaptation in order to achieve the target that was compressed by the average population studied. Seeks to that this project makes an improvement in quality of life of people in wheelchairs by including these people in society but also the improvement of rehabilitation

Gestão da satisfação e fidelidade do cliente : um estudo dos fatores que influenciam na satisfação e fidelidade dos clientes corporativos de telefonia celular

The competition in the telecommunications industry has grown in Brazil since the privatization, forcing companies that are active in the market to a growing commitment to quality products and services in order to survive. In this context, this work aims to understand the main factors that influence the degree of satisfaction exists in respect of a mobile operator with its corporate customers. The research covered theoretical concepts and analytical models of quality management system and models of indices related to the measurement of customer satisfaction. For the field research was carried out in a practical application of the main approaches based on this thesis by a case study in corporate segment, through a questionnaire applied to 10 consultants and 40 corporate customers of that company. Comparing the results of research with the consultants and corporate clients there is the concern of respondents to the indicators that comprise the constructs of customer satisfaction, commitment calculated, the price index and the handling of complaints, denoting the dissatisfaction of the general assessment for corporate customers with the carrier, against its current expectations. It is concluded that the mobile operator of the telecommunications industry have a big challenge, after ten years of privatization and consequently the period of rapid expansion of customer base and with the depleted, retain corporate customers as highly strategic, thus avoiding that migrate to other companies. We emphasize the need for further research and analysis of different approaches through research and using the same models to specifically evaluate and measure customer satisfaction of mobile enterprise, to adjust the model to the national market. Finally, we suggest the creation of an effective customer loyalty program with a strategy of relationship and specific to the corporate sector of mobile telephony

Estudo de sistemas complexos com interações de longo alcance : percolação, redes e tráfego

In this thesis we investigate physical problems which present a high degree of complexity using tools and models of Statistical Mechanics. We give a special attention to systems with long-range interactions, such as one-dimensional long-range bondpercolation, complex networks without metric and vehicular traffic. The flux in linear chain (percolation) with bond between first neighbor only happens if pc = 1, but when we consider long-range interactions , the situation is completely different, i.e., the transitions between the percolating phase and non-percolating phase happens for pc < 1. This kind of transition happens even when the system is diluted ( dilution of sites ). Some of these effects are investigated in this work, for example, the extensivity of the system, the relation between critical properties and the dilution, etc. In particular we show that the dilution does not change the universality of the system. In another work, we analyze the implications of using a power law quality distribution for vertices in the growth dynamics of a network studied by Bianconi and Barabási. It incorporates in the preferential attachment the different ability (fitness) of the nodes to compete for links. Finally, we study the vehicular traffic on road networks when it is submitted to an increasing flux of cars. In this way, we develop two models which enable the analysis of the total flux on each road as well as the flux leaving the system and the behavior of the total number of congested roads

Administração repetida de baixas doses de reserpina: um possível modelo para o estudo de déficits cognitivos e motores associados à Doença de Parkinson

Parkinson's disease (PD) is one of the most common neurodegenerative brain disorders and is characterized primarily by a progressive degeneration of dopaminergic neurons nigroestriatais. The main symptoms of this disease are motor alterations (bradykinesia, rigidity, tremor at rest), which can be highly disabling in advanced stages of the condition. However, there are symptomatic manifestations other than motor impairment, such as changes in cognition, mood and sensory systems. Animal models that attempt to mimic clinical features of PD have been used to understand the behavioral and neural mechanisms underlying neurophysiological disturbance of this disease. However, most models promote an intense and immediate motor impairment, consistent with advanced stages of the disease, invalidating these studies for the evaluation of its progressive nature. The administration of reserpine (a monoamine depletor) in rodents has been considered an animal model for studying PD. Recently we found that reserpine (in doses lower than those usually employed to produce the motor symptoms) promotes a memory deficit in an aversive discrimination task, without changing the motor activity. It was suggested that the administration of this drug in low doses can be useful for the study of memory deficits found in PD. Corroborating this data, in another study, acute subcutaneous administration of reserpine, while preserving motor function, led to changes in emotional context-related (but not neutral) memory tasks. The goal of this research was to study the cognitive and motor deficits in rats repeatedly treated with low doses of reserpine, as a possible model that simulates the progressive nature of the PD. For this purpose, 5-month-old male Wistar rats were submitted to a repeated treatment with vehicle or different doses of reserpine on alternate days. Cognitive and motor parameters and possible changes in neuronal function were evaluated during treatment. The main findings were: repeated administration of 0.1 mg / kg of reserpine in rats is able to induce the gradual appearance of motor signs compatible with progressive features found in patients with PD; an increase in striatal levels of oxidative stress and changes in the concentrations of glutamate in the striatum were observed five days after the end of treatment; in animals repeatedly-treated with 0. 1 mg/kg, cognitive deficits were observed only after the onset of motor symptoms, but not prior to the onset of these symptoms; 0.2 mg / kg reserpine repeated treatment has jeopardized the cognitive assessment due to the presence of severe motor deficits. Thus, we suggest that the protocol of treatment with reserpine used in this work is a viable alternative for studies of the progressive appearance of parkinsonian signs in rats, especially concerning motor symptoms. As for the cognitive symptoms, we suggest that more studies are needed, possibly using other behavioral models, and / or changing the treatment regimen