A dinâmica da taxa de câmbio no Brasil face às operações swap (2002-2015)

This dissertation investigates, based on the Post-Keynesian theory and on its concept of monetary economy of production, the exchange rate behavior of the Brazilian Real in the presence of Brazilian Central Bank's interventions by means of the so-called swap transactions over 2002-2015. Initially, the work analyzes the essential properties of an open monetary economy of production and, thereafter, it presents the basic propositions of the Post-Keynesian view on the exchange rate determination, highlighting the properties of foreign exchange markets and the peculiarities of the Brazilian position into the international monetary and financial system. The research, thereby, accounts for the various segments of the Brazilian foreign exchange market. To accomplish its purpose, we first do a literature review of the Post-Keynesian literature about the topic. Then, we undertake empirical exams of the exchange rate determination using two statistical methods. On the one hand, to measure the volatility of exchange rate, we estimate Auto-regressive Conditional Heteroscedastic (ARCH) and Generalized Auto-regressive Conditional Heteroscedastic (GARCH) models. On the other hand, to measure the variance of the exchange rate in relation to real, financial variables, and the swaps, we estimate a Vector Auto-regression (VAR) model. Both experiments are performed for the nominal and real effective exchange rates. The results show that the swaps respond to exchange rate movements, trying to offset its volatility. This reveals that the exchange rate is, at least in a certain magnitude, sensitive to swaps transactions conducted by the Central Bank. In addition, another empirical result is that the real effective exchange rate responds more to the swaps auctions than the nominal rate.

Determinaçao das principais proteínas de fase aguda e do índice prognóstico inflamatório nutricional (IPIN) em cachorrodo-mato (Cerdocyon thous - Linnaeus, 1766)

The dog-eating fox (Cerdocyon thous - Linnaeus, 1766) is a medium sized canid widely distributed in South America and occurs in almost all of Brazil. Among the main threats to their conservation are the roadkill mainly caused by habitat loss. The shortage of laboratory bush dogs data affect the veterinary medical care hindering the application of appropriate therapies. This study aimed to evaluate the levels of C-reactive protein, albumin, pre-albumin, ceruloplasmin, haptoglobin and Afla 1 acid glycoprotein and the Prognostic Index Inflammatory Nutritional (IPIN) in this species, thus obtaining a first description of these prognostic markers. They collected 1.5 ml of blood by jugular access 8 of Mato Dogs copies (thous thous) from the Laboratory of collection of Teaching and Research in Wildlife (limpets), Faculty of Veterinary Medicine, Federal University of Uberlândia for exams routine. The samples were collected via the jugular vein after physical restraint of animals and trichotomy of the region. After statistical analysis, the values were: albumin: between 2.7 and 3.0 g / dl, alpha 1-acid glycoprotein: between 0.19 and 0.21 g / l, C-reactive protein: between 1.7 and 2 2, prealbumin between 30 and 35 mg / l haptoglobin: between 0.078 and 0.156 and IPIN ≤ 0.006 being considered normal and values ≥ 0.006 considered high. This press description will serve as a basis for studies where animals may be used with specific diseases and, after analysis, compared with the values found in this study and verified the behavior follows the likeness of domestic dogs.

Classificação de nódulos pulmonares em imagens tomográficas utilizando redes neurais artificiais em cascata

Lung cancer is the most common of malignant tumors, with 1.59 million new cases worldwide in 2012. Early detection is the main factor to determine the survival of patients affected by this disease. Furthermore, the correct classification is important to define the most appropriate therapeutic approach as well as suggest the prognosis and the clinical disease evolution. Among the exams used to detect lung cancer, computed tomography have been the most indicated. However, CT images are naturally complex and even experts medical are subject to fault detection or classification. In order to assist the detection of malignant tumors, computer-aided diagnosis systems have been developed to aid reduce the amount of false positives biopsies. In this work it was developed an automatic classification system of pulmonary nodules on CT images by using Artificial Neural Networks. Morphological, texture and intensity attributes were extracted from lung nodules cut tomographic images using elliptical regions of interest that they were subsequently segmented by Otsu method. These features were selected through statistical tests that compare populations (T test of Student and U test of Mann-Whitney); from which it originated a ranking. The features after selected, were inserted in Artificial Neural Networks (backpropagation) to compose two types of classification; one to classify nodules in malignant and benign (network 1); and another to classify two types of malignancies (network 2); featuring a cascade classifier. The best networks were associated and its performance was measured by the area under the ROC curve, where the network 1 and network 2 achieved performance equal to 0.901 and 0.892 respectively.