Risco de crédito: uma abordagem utilizando análise discriminante, regressão logística e redes neurais artificiais

Considering the relevance of researches concerning credit risk, model diversity and the existent indicators, this thesis aimed at verifying if the Fleuriet Model contributes in discriminating Brazilian open capital companies in the analysis of credit concession. We specifically intended to i) identify the economic-financial indicators used in credit risk models; ii) identify which economic-financial indicators best discriminate companies in the analysis of credit concession; iii) assess which techniques used (discriminant analysis, logistic regression and neural networks) present the best accuracy to predict company bankruptcy. To do this, the theoretical background approached the concepts of financial analysis, which introduced themes relative to the company evaluation process; considerations on credit, risk and analysis; Fleuriet Model and its indicators, and, finally, presented the techniques for credit analysis based on discriminant analysis, logistic regression and artificial neural networks. Methodologically, the research was defined as quantitative, regarding its nature, and explanatory, regarding its type. It was developed using data derived from bibliographic and document analysis. The financial demonstrations were collected by means of the Economática ® and the BM$FBOVESPA website. The sample was comprised of 121 companies, being those 70 solvents and 51 insolvents from various sectors. In the analyses, we used 22 indicators of the Traditional Model and 13 of the Fleuriet Model, totalizing 35 indicators. The economic-financial indicators which were a part of, at least, one of the three final models were: X1 (Working Capital over Assets), X3 (NCG over Assets), X4 (NCG over Net Revenue), X8 (Type of Financial Structure), X9 (Net Thermometer), X16 (Net Equity divided by the total demandable), X17 (Asset Turnover), X20 (Net Equity Profitability), X25 (Net Margin), X28 (Debt Composition) and X31 (Net Equity over Asset). The final models presented setting values of: 90.9% (discriminant analysis); 90.9% (logistic regression) and 97.8% (neural networks). The modeling in neural networks presented higher accuracy, which was confirmed by the ROC curve. In conclusion, the indicators of the Fleuriet Model presented relevant results for the research of credit risk, especially if modeled by neural networks.

Nanocompósitos com atividade antioxidante de amido de milho/isolado proteico de soro de leite/ óleo essencial de alecrim obtidos por extrusão termoplástica

In order to improve the quality and safety of food, the active packaging emerges as a new technology based on the release of composites beneficial to food products. Thus, biodegradable films incorporated with active substances have the function of acting as a barrier to external elements, protecting the product and increasing its shelf life. They are formulated from proteins, polysaccharides, lipids or from the combination of these compounds. However, there is a need to improve the performance properties of these packages. Nanotechnologies, then, emerges with the study of many nanoparticles as additives to modify the performance of biodegradable polymers. With this, we aimed at developing and active antioxidant film of corn starch blenders and whey protein isolate with rosemary essential oil or microcapsules of rosemary essential oil reinforced with sodium montmorillonite (MMTNa + ) nanoparticles by extrusion. The films were developed and characterized in a first stage for the selection of the best polymeric blender using the following analyses: water vapor permeability (WVP), machanical properties; optical, thermogravimetry (TG), differential scanning calorimetry (DSC), x-ray diffraction (XRD) and scanning electron microscopy (SEM). In the second stage, montmorillonite clay nanoparticles and rosemary essential oil were added as reinforcement to evaluate its antioxidant effect. In a third stage, we studied the addition of microcapsules of rosemary essential oil (MR) as a form of protecting the active agent and its antioxidant potential in the films. The results indicate that the development of p olymeric blender with 30% of corn starch substitution is the most indicated for future work. The addition of rosemary essential oil or microcapsule of rosemary essential oil allowed for the obtaining of nanocomposites with antioxidant potential for application in food packages.

Inovação do método MIA-QSAR: aplicação para doenças tropicais negligenciadas

Multivariate image analysis applied to the quantitative structure-activity relationships (MIA-QSAR) is a 2D QSAR technique that has been presenting promising outcomes for the development of new drug candidates, due to its simplicity, rapidity and low cost. In this way, the present study aims at introducing, consolidating and improving the new dimensions named aug-MIA-QSAR and aug-MIA-QSARcolor, as well as applying them to the study of neglected diseases, in order to obtain new drug targets using chemico-biological interpretation of the MIA molecular descriptors. Four compound data sets with experimental bioactivities against Chagas disease, malaria, dengue and schistosomiasis were evaluated using three approaches: MIA-QSARt, aug-MIA-QSAR and aug-MIA-QSARcolor. In general, representations of atoms as spheres with different colors and sizes proportional to the corresponding van der Waals radii (aug-MIA approaches) improved the predictive ability and interpretability in all data sets. The use of colors proportional to the Pauling´s electronegativity showed that MIA descriptors are capable of identifying periodic properties relevant for the studied activity. Finally, solid colors instead of spotlighted atoms allowed a correct identification of atoms by means of pixel values in the studies for malaria, dengue and schistosomiasis, which were, subsequently, useful for the chemical interpretation related to the bioactivity. It can be inferred that semicarbazones and thiosemicarbazones derivative with a tri-substituted ring in R1 group and a trifluoro methyl group in the R 3 position instead of a chlorine antitripanossoma resulted in higher activity. The antimalarial activity of quinolon-4(1H)imines can be improved if: 1) R1 and R2 are electron donor groups, 2) R3 has long aminoalkyl chains, and 3) R4 possesses substituents with big atomic volume. In the study for dengue, it was found that tetrapeptides with unbranched small size amino acids in the A1 and A4 positions can increase the substrate affinity (Km) to the NS3 protein, and when in A1 and A2 positions, the substrate cleavage rate (kcat). On the other hand, acidic amino acids in the A2 and A4 positions were found to be related with low substrate affinity to the NS3 protein and when present in A1, with low substrate cleavage rate. Finally, the presence of metoxy substituents in R1 (or R2) and R5 in the neolignan backbone can favor their antischistosomal activity.