Monitoramento de processos multivariados autocorrelacionados

Technological advances and the availability of computational resources have been facilitating the collection and processing of data. Thus, the natural tendency of the monitoring processes is the simultaneous control of various quality characteristics. In automated processes, observations are generally autocorrelated. Studies with univariate graph for processes have shown that the autocorrelation reduces the ability of this signal changes in the process. In this paper, we study the multivariate autocorrelated processes. Through simulations are obtained properties of graphs, monitoring the mean vector, the properties of graphs VMAX, in monitoring the covariance matrix, and the properties of graphs MCMAX, the simultaneous monitoring of mean vector and covariance matrix. Conclude that increasing the autocorrelation and the number of variables being monitored, reduces the power of the graphics in signal of a special cause

Uma discussão sobre o ensino da estatística para os estudantes de graduação na França, no Brasil, e nos Estados Unidos

Today statistics are one of the most important bases of the engineer formation. Statistics are used indirectly every day for engineers in a large panel of sectors like construction, mechanical engineering, biological engineering, electrical engineering, computer science etc. The main goal of this report is to compare different ways the engineers of tomorrow are formed in order to show which engineering is the best practice. The comparison will be done on different universities in France, Brazil and the US on Mechanical, Civil, Electrical and Production branches. We will compare the amount of courses required, the degree of knowledge needed and the mains subjects

Estudo do método Monte Carlo e sua aplicação no modelo SIR

In this work we present a discussion and the results of the simulation of disease spread using the Monte Carlo method. The dissemination model is the SIR model and presents as main characteristic the disease evolution among individuals of the population subdivided into three groups: susceptible (S), infected (I) and recovered (R). The technique used is based on the introduction of transition probabilities S-> I and I->R to do the spread of the disease, they are governed by a Poisson distribution. The simulation of the spread of disease was based on the randomness introduced, taking into account two basic parameters of the model, the power of infection and average time of the disease. Considering appropriate values of these parameters, the results are presented graphically and analysis of these results gives information on a group of individuals react to the changes of these parameters and what are the chances of a disease becoming a pandemic

Sincronia das bolsas de valores

This study aims to analyze various index stock exchanges around the world, with mathematical methods, analyzing the Product Distribution function (PDF), cumulative distribution and correlation. It is thought that the world economy is connected, as if the financial markets were a network, where the fluctuation of a market generates a variation on another and another, creating a pattern, that this change will affect the entire network, thus creating what we might call the domino effect. From this we intend to study, using as a basis the main index of the major stock exchanges around the world the relationship they have with each other, analyzing the influence and correlation that generates this effect, showing that markets are connected and influence each other. We can see this effect in the crisis of 2008, where the U.S. market from one moment to the other was shaken, affecting the whole world in a few days, creating effects that are felt in the present day

Caracterização da sensibilidade ambiental da Baixada Santista no estado de São Paulo

This study aimed to perform the analysis and characterization of environmental sensitivity to oil from Baixada Santista in the State of São Paulo. The work was done by integrating data of physical environmet, socio-economic activities and the biological / ecological provided by the Research Group on Environmental Sensitivity to Oil Spill in the Institute of Geosciences and Exact Sciences of UNESP “Julio de Mesquita Filho” wich works in the conjunction with the Program of Human Resources Training in Geosciences and Environmental Sciences Applied to the Oil and Gas (PRH-05) of the National Petroleum Agency (ANP). Were also performed descriptive statistical analysis, based on dispersion and trend parameters, which allowed to answer questions related to environmental sensitivity index (ISL) prevalence, the predominance of the ISL by environment and the predominance of the environments of the area, thus providing an overview of Baixada Santista’s main towns about sensitivity to oil. Analyses performed in this study may also help mitigate the environmental and socioeconomic impacts and contribute to contingency plans development for Baixada Santista

O efeito da autocorrelação no desempenho dos gráficos de X e R e dos gráficos de X e S²

Control charts are very important tools in statistical quality control of industrial processes and its use started last century. Since its development, the charts have always been attributed to independent processes, i.e. without any correlation between samples. But nowadays, with the high level of automation in the industrial environment, it is noticeable the autocorrelation factor between samples. The main Xcharts used in monitoring quality characteristics represented by continuous variables are the mean (X ), amplitude (R) and variance (S²). Therefore, this work aims to analyze the performance of X and R charts and in of X and S² charts with different sample sizes (4 and 5) for monitoring autocorrelated processes. Through computer simulations using the Fortran software and the use of mathematical expressions was possible to obtain data and performance analysis of the detection power charts for independent observations and for autocorrelated observations according to the model AR (1). The results show that the effect of autocorrelation reduces the ability of monitoring the control charts and that, the greater this effect, the slower the chart becomes in misfits signaling

Técnicas de análise multivariada com aplicações a dados de natureza biológica

A análise estatística multivariada, extensão da análise univariada, consiste num conjunto de técnicas estatísticas, aplicadas quando há diversas variáveis relacionadas simultaneamente, sendo todas elas, em princípio, consideradas importantes no fenômeno em estudo. É de grande aplicação a conjuntos de dados das mais diversas áreas do conhecimento, principalmente da área biológica. Seu desenvolvimento teve um grande impulso na primeira metade do século passado. Entretanto, devido a complexidade dos cálculos matemáticos, principalmente envolvendo operações com matrizes de altas ordens, as aplicações somente se popularizaram nos dias atuais, com o desenvolvimento dos computadores e aplicativos computacionais. Técnicas estudadas: distâncias multivariadas, componentes principais, análise fatorial, correlações canônicas, análise de correspondência, teste t² de Hotelling, análise de variância multivariada (Manova), teste de normalidade multivariada, igualdade de matrizes de variâncias e covariâncias para populações multinormais

Análise histomorfométrica do músculo cardíaco de ratos recém natos gerados por matrizes submetidas ao tabagismo passivo associado ao exercício

Maternal smoking during pregnancy causes heart changes, impaired fetal growth, spontaneous abortion and low birth weight. Even in pregnant women exposed to cigarette smoke these changes can be observed. But the exercise is to answer the increased fetal weight, besides the improvement of cardiorespiratory fitness of the pregnant and the fetus. The objective of this study was to analyze the ventricular muscle of young rats subjected to passive smoking associated with aquatic exercise. For the experiments used 24 female rats were divided into: GC (control), GE (exercise protocol in the water), GF(exposed to cigarette smoke) and GEF (exercise protocol on water and exposed to cigarette smoke). On the same day was held the 1st session of the experimental phase of the protocol of exposure to cigarette smoke, which consisted of 30 minutes twice a day, six days a week for three weeks, followed by the first session of the swimming program... (Complete abstract click electronic access below)

Controle da população de morcegos hematófagos na região de Andradina, São Paulo

Pós-graduação em Ciência Animal - FMVA

Relação solo-vegetação em área de cerrado no nordeste do Maranhão, Brasil

Pós-graduação em Agronomia (Ciência do Solo) - FCAV

Modelagem espacial de áreas de recarga subterrânea em região de afloramento do sistema aquífero guarani (sag), em Brotas/SP

Pós-graduação em Agronomia (Irrigação e Drenagem) - FCA

Aplicação da análise fatorial em cidades do Vale do Paraíba considerando os objetivos do milênio

Pós-graduação em Engenharia Mecânica - FEG

Genetic analysis of feet and legs in Nelore cattle

Pós-graduação em Genética e Melhoramento Animal - FCAV

Parâmetros genéticos para características de desempenho e reprodutivas de aves poedeiras por inferência bayesiana

Pós-graduação em Genética e Melhoramento Animal - FCAV

Gráfico de Hotelling com esquemas especiais de amostragem para o monitoramento de processos bivariados autocorrelacionados

Pós-graduação em Engenharia Mecânica - FEG