Métodos lineares e não lineares de análise de séries temporais e sua aplicação no estudo da variabilidade da frequência cardíaca de jovens saudáveis

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Análise comportamental dos processos de ensino e de aprendizagem de conteúdos curriculares de língua portuguesa nas séries iniciais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Análise bayesiana do modelo fatorial dinâmico para um vetor de séries temporais utilizando a distribuição t multivariada

The multivariate t models are symmetric and with heavier tail than the normal distribution, important feature in financial data. In this theses is presented the Bayesian estimation of a dynamic factor model, where the factors follow a multivariate autoregressive model, using multivariate t distribution. Since the multivariate t distribution is complex, it was represented in this work as a mix between a multivariate normal distribution and a square root of a chi-square distribution. This method allowed to define the posteriors. The inference on the parameters was made taking a sample of the posterior distribution, through the Gibbs Sampler. The convergence was verified through graphical analysis and the convergence tests Geweke (1992) and Raftery & Lewis (1992a). The method was applied in simulated data and in the indexes of the major stock exchanges in the world.

O processo de municipalização do ensino no estado de São Paulo: uma análise do seu desenvolvimento na região de governo de Marília

Pós-graduação em Educação - FFC

Análise de séries temporais de multicaminho em estações de monitoramento contínuo

Data from reference stations are widely used in GNSS (Global Navigation Satellite System) positioning, and can be used in relative positioning or network-based positioning concept. Positioning accuracy will be directly influenced by errors in signals collected in these stations. In this paper, it is aimed at evaluating these data quality using temporal series of multipath index MP1 and MP2. A statistical study of temporal series with 7 years of daily observations related to 7 stations from RBMC (Rede Brasileira de Monitoramento Contínuo) was accomplished. In order to investigate trends and seasonality a linear regression model, correlograms, and Fourier periodograms were used. We also used a harmonic adjust to identify peaks on temporal series. At last, the possible causes of seasonality found in some stations were discussed. It was also possible to identify peaks in MP values of March and October months (mainly in stations located near geomagnetic equator).

Análise multiescala de séries temporais do efeito da cintilação ionosférica nos sinais de satélite GPS a partir de wavelets não decimadas

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

O processo de configuração curricular: uma análise da compreensão de professores de educação física das séries iniciais do ensino fundamental

Pós-graduação em Educação - IBRC

Aplicação da análise de séries temporais para detecção e prognóstico de danos em estruturas inteligentes

Pós-graduação em Engenharia Mecânica - FEIS

Geotecnologias aplicadas à análise do assoreamento de reservatórios de hidrelétricas em cenário de escassez de dados de séries históricas

Pós-graduação em Engenharia Mecânica - FEG

Análise de variância WAVELET de séries temporais com dados faltantes: avaliação do índice de cintilação ionosférica S4 em sinais de satélite GPS

Pós-graduação em Matematica Aplicada e Computacional - FCT

Análise de variância WAVELET de séries temporais com dados faltantes: avaliação do índice de cintilação ionosférica S4 em sinais de satélite GPS

Pós-graduação em Matematica Aplicada e Computacional - FCT

Análise de métodos para detecção de mudanças estruturais em séries temporais climáticas

The time series analysis has played an increasingly important role in weather and climate studies. The success of these studies depends crucially on the knowledge of the quality of climate data such as, for instance, air temperature and rainfall data. For this reason, one of the main challenges for the researchers in this field is to obtain homogeneous series. A time series of climate data is considered homogeneous when the values of the observed data can change only due to climatic factors, i.e., without any interference from external non-climatic factors. Such non-climatic factors may produce undesirable effects in the time series, as unrealistic homogeneity breaks, trends and jumps. In the present work it was investigated climatic time series for the city of Natal, RN, namely air temperature and rainfall time series, for the period spanning from 1961 to 2012. The main purpose was to carry out an analysis in order to check the occurrence of homogeneity breaks or trends in the series under investigation. To this purpose, it was applied some basic statistical procedures, such as normality and independence tests. The occurrence of trends was investigated by linear regression analysis, as well as by the Spearman and Mann-Kendall tests. The homogeneity was investigated by the SNHT, as well as by the Easterling-Peterson and Mann-Whitney-Pettit tests. Analyzes with respect to normality showed divergence in their results. The von Neumann ratio test showed that in the case of the air temperature series the data are not independent and identically distributed (iid), whereas for the rainfall series the data are iid. According to the applied testings, both series display trends. The mean air temperature series displays an increasing trend, whereas the rainfall series shows an decreasing trend. Finally, the homogeneity tests revealed that all series under investigations present inhomogeneities, although they breaks depend on the applied test. In summary, the results showed that the chosen techniques may be applied in order to verify how well the studied time series are characterized. Therefore, these results should be used as a guide for further investigations about the statistical climatology of Natal or even of any other place.

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