Uso de modelos estatísticos para dados de escore de crédito de uma instituição financeira

Pós-graduação em Engenharia de Produção - FEB

Specialist tool for monitoring the measurement degradation process of induction active energy meters

This paper presents a methodology and a specialist tool for failure probability analysis of induction type watt-hour meters, considering the main variables related to their measurement degradation processes. The database of the metering park of a distribution company, named Elektro Electricity and Services Co., was used for determining the most relevant variables and to feed the data in the software. The modeling developed to calculate the watt-hour meters probability of failure was implemented in a tool through a user friendly platform, written in Delphi language. Among the main features of this tool are: analysis of probability of failure by risk range; geographical localization of the meters in the metering park, and automatic sampling of induction type watt-hour meters, based on a risk classification expert system, in order to obtain information to aid the management of these meters. The main goals of the specialist tool are following and managing the measurement degradation, maintenance and replacement processes for induction watt-hour meters. © 2011 IEEE.

Mercado de ações brasileiro: uma investigação empírica sobre suas relações de longo prazo e de precedência temporal pré-crise de 2008

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

CHARACTERIZATION OF THE BRAZPD II COHORT AND DESCRIPTION OF TRENDS IN PERITONEAL DIALYSIS OUTCOME ACROSS TIME PERIODS

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Peritoneal dialysis in acute kidney injury: trends in the outcome across time periods

Peritoneal dialysis (PD) should be considered a suitable method of renal replacement therapy in acute kidney injury (AKI) patients. This study is the largest cohort providing patient characteristics, clinical practice, patterns and their relationship to outcomes in a developing country. Its objective was to describe the main determinants of patient and technique survival, including trends over time of PD treatment in AKI patients. This was a Brazilian prospective cohort study in which all adult AKI patients on PD were studied from January/2004 to January/2014. For comparison purposes, patients were divided into 2 groups according to the year of treatment: 2004-2008 and 2009-2014. Patient survival and technique failure (TF) were analyzed using the competing risk model of Fine and Gray. A total of 301 patients were included, 51 were transferred to hemodialysis (16.9%) during the study period. The main cause of TF was mechanical complication (47%) followed by peritonitis (41.2%). There was change in TF during the study period: compared to 2004-2008, patients treated at 2009-2014 had relative risk (RR) reduction of 0.86 (95% CI 0.77-0.96) and three independent risk factors were identified: period of treatment at 2009 and 2014, sepsis and age>65 years. There were 180 deaths (59.8%) during the study. Death was the leading cause of dropout (77.9% of all cases) mainly by sepsis (58.3%), followed cardiovascular disease (36.1%). The overall patient survival was 41% at 30 days. Patient survival improved along study periods: compared to 2004-2008, patients treated at 2009-2014 had a RR reduction of 0.87 (95% CI 0.79-0.98). The independent risk factors for mortality were sepsis, age>70 years, ATN-ISS > 0.65 and positive fluid balance. As conclusion, we observed an improvement in patient survival and TF along the years even after correction for several confounders and using a competing risk approach.

Influence of a probiotic soy product on fecal microbiota and its association with cardiovascular risk factors in an animal model

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

Establishing the risk of neonatal mortality using a fuzzy predictive model

O objetivo do artigo foi avaliar o uso da lógica fuzzy para estimar possibilidade de óbito neonatal. Desenvolveu-se um modelo computacional com base na teoria dos conjuntos fuzzy, tendo como variáveis peso ao nascer, idade gestacional, escore de Apgar e relato de natimorto. Empregou-se o método de inferência de Mamdani, e a variável de saída foi o risco de morte neonatal. Criaram-se 24 regras de acordo com as variáveis de entrada, e a validação do modelo utilizou um banco de dados real de uma cidade brasileira. A acurácia foi estimada pela curva ROC; os riscos foram comparados pelo teste t de Student. O programa MATLAB 6.5 foi usado para construir o modelo. Os riscos médios foram menores para os que sobreviveram (p < 0,001). A acurácia do modelo foi 0,90. A maior acurácia foi com possibilidade de risco igual ou menor que 25% (sensibilidade = 0,70, especificidade = 0,98, valor preditivo negativo = 0,99 e valor preditivo positivo = 0,22). O modelo mostrou acurácia e valor preditivo negativo bons, podendo ser utilizado em hospitais gerais.

Influence of a probiotic soy product on fecal microbiota and its association with cardiovascular risk factors in a animal model

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