No abstract

In this work we study the influence of the Sobradinho dam construction on daily streamflow of São Francisco River, Brasil, by analyzing long-range correlations in magnitude and sign time series obtained from streamflow anomalies, using the Detrended Fluctuation Analysis (DFA) method. The magnitude series relates to the nonlinear properties of the original time series, while the sign series relates to the linear properties. The streamflow data recorded during the period 1929-2009, were divided in the periods pre-construction (1929 to 1972) and post-construction (1980 to 2009) of Sobradinho dam and analyzed for small scales (less than 1 year) and for large scales (more than 1 year). In post-construction of Sobradinho dam, DFA-exponents of magnitude series increased at small scales (0.895 to 1.013) and at large scales (0.371 to 0.619) indicating that the memory associated with nonlinear components becames stronger. For sign series, the DFA-exponent increased at small scales (0.596 to 0.692) indicating stronger persistence of flow increments direction, and decreased at large scales (0.381 to 0.259) indicating stronger anti-persistence (positive increments are more likely to be followed by negative increments and vice versa). These results provide new evidence on the hydrological changes in the São Francisco River caused by human activities.

Atualmente a análise de sobrevivência é uma das áreas que mais crescem no campo da análise estatística, com uma sólida teoria para ajustar modelos de regressão para estudar certos fenômenos, os quais têm, em sua estrutura, a característica de ter observações incompletas na amostra denominada censura. Embora esses modelos possam representar eficientemente o fenômeno em estudo em muitas situações, alguns deles não levam em consideração a existência de uma variável não observável presente na maioria dos estudos, denominada fragilidade. Essa fragilidade denota a suscetibilidade do evento a ocorrer por um indivíduo ou objeto determinado sob investigação. O objetivo deste trabalho foi mostrar que, em situações em que a fragilidade está presente, o uso de modelos que capturam a variabilidade dessa variável é mais viável para a análise desses dados quando comparado aos modelos convencionais em estudos de sobrevivência. Para tanto, foi realizada uma análise comparativa entre esses modelos, ajustada para um conjunto de dados de pacientes diagnosticados com retinopatia diabética, e também foi realizado um estudo de simulação para o modelo de fragilidade gama com diferentes porcentagens de censura e heterogeneidade. Após o ajuste dos modelos, observa-se que os modelos de fragilidade tiveram melhor desempenho quando comparados ao modelo de Cox, com ênfase no modelo de fragilidade gama, que gerou o menor valor para AIC e BIC. O estudo de simulação mostrou que altas taxas de censura prejudicam o grau de previsibilidade do modelo de fragilidade e que altas taxas de heterogeneidade contribuem para estimativas de parâmetros.

This study evaluates the Brazilian agricultural commodities market and the dollar–real exchange price variation using the multifractal detrended fluctuations analysis methodology. We investigated the period from January 1, 2019 to September 25, 2019, outside the COVID-19 pandemic, and from January 1, 2020 to September 25, 2020, during the COVID-19 pandemic. We verified the fluctuations of commodities and dollar–real exchange prices during the pandemic caused by COVID-19 showed a record price. The results of Hurst exponent and multifractal parameters [Formula: see text], [Formula: see text], and [Formula: see text] indicate that during the COVID-19 pandemic, sugar was the most efficient commodity, while pork the less one. Compared to the identical months in 2019, the dollar–real exchange was the most efficient market, while ethanol was the least efficient.

The Agriculture sector has created and collected large amounts of data. It can be gathered, stored, and analyzed to assist in decision making generating competitive value, and the use of Machine Learning techniques has been very effective for this market. In this work, a Machine Learning study was carried out using supervised classification models based on boosting to predict disease in a crop, thus identifying the model with the best areas under curve metrics. Light Gradient Boosting Machine, CatBoost Classifier, Extreme Gradient, Gradient Boosting Classifier, Adaboost models were used to qualify the crop as healthy or sick. One can see that the LightGBM algorithm provided a better fit to the data with an area under the curve of 0.76 under the use of BORUTA variable selection.