17 resultados para weighted linear regression
em Repositório Institucional UNESP - Universidade Estadual Paulista "Julio de Mesquita Filho"
Resumo:
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
Resumo:
We consider model selection uncertainty in linear regression. We study theoretically and by simulation the approach of Buckland and co-workers, who proposed estimating a parameter common to all models under study by taking a weighted average over the models, using weights obtained from information criteria or the bootstrap. This approach is compared with the usual approach in which the 'best' model is used, and with Bayesian model averaging. The weighted predictor behaves similarly to model averaging, with generally more realistic mean-squared errors than the usual model-selection-based estimator.
Resumo:
Este trabalho teve por objetivo estimar equações de regressão linear múltipla tendo, como variáveis explicativas, as demais características avaliadas em experimento de milho e, como variáveis principais, a diferença mínima significativa em percentagem da média (DMS%) e quadrado médio do erro (QMe), para peso de grãos. Com 610 experimentos conduzidos na Rede de Ensaios Nacionais de Competição de Cultivares de Milho, realizados entre 1986 e 1996 (522 experimentos) e em 1997 (88 experimentos), estimaram-se duas equações de regressão, com os 522 experimentos, validando estas pela análise de regressão simples entre os valores reais e os estimados pelas equações, com os 88 restantes, observando que, para a DMS% a equação não estimava o mesmo valor que a fórmula original e, para o QMe, a equação poderia ser utilizada na estimação. Com o teste de Lilliefors, verificou-se que os valores do QMe aderiam à distribuição normal padrão e foi construída uma tabela de classificação dos valores do QMe, baseada nos valores observados na análise da variância dos experimentos e nos estimados pela equação de regressão.
Resumo:
Em geral, a função de um modelo de impedância para processos de eletrodo simples é deduzida de um modelo elétrico equivalente, denominado circuito de Randles. Neste trabalho estudou-se a generalização dessa função, mediante a introdução de um parâmetro não-elétrico, relacionado com a flexibilidade do ângulo de fase e da magnitude. A função foi ajustada às medidas experimentais de impedância obtidas de um sistema constituído de uma liga Ti-10%Al (m/m) em solução de cloreto de sódio 0,9%, variando-se a amplitude de perturbação. Verificou-se que a função generalizada foi adequada para descrever a impedância do sistema analisado, reduzindo as distorções entre a curva experimental e a curva teórica. Além disso, os melhores resultados foram obtidos para sinais de perturbação do sistema com amplitude igual a 10 mV.
Resumo:
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
Resumo:
O fitotensiômetro de Alvim é amplamente utilizado para demonstrar a variação da turgescência do tronco, que é reduzida durante o dia, como efeito líquido da transpiração que ocorre na copa, e aumentada durante o período noturno, devido à recuperação que pode ser total ou parcial, dependendo da umidade disponível no solo, e da capacidade da planta em absorvê-la. Os valores indicados no manômetro são relativos, e muitas vezes não coincidem com os previstos pelas taxas de transpiração mensuradas ao longo do dia. Este trabalho considera a necessidade de uma nova interpretação das curvas observadas, através do seu seccionamento em diferentes cinéticas, sendo aplicadas equações de regressão linear a estas diferentes partes, e analisado o coeficiente angular destas regressões como parâmetro comparativo da intensidade das mudanças na turgescência do tronco entre diferentes plantas, sob diferentes condições ambientais.
Resumo:
Globalization of dairy cattle breeding has created a need for international sire proofs. Some early methods for converting proofs from one population to another are based on simple linear regression. An alternative robust regression method based on the t-distribution is presented, and maximum likelihood and Bayesian techniques for analysis are described, including the situation in which some proofs are missing. Procedures were used to investigate the relationship between Holstein sire proofs obtained by two Uruguayan genetic evaluation programs. The results suggest that conversion equations developed from data including only sires having proofs in both populations can lead to distorted results, relative to estimates obtained using techniques for incomplete data. There was evidence of non-normality of regression residuals, which constitutes an additional source of bias. A robust estimator may not solve all problems, but can provide simple conversion equations that are less sensitive to outlying proofs and to departures from assumptions.
Resumo:
Objective. To assess factors determining growth in a group of children between 3 months and 6 years old enrolled in a public municipal (i.e., government-supported, not private) day-care center, in comparison to a group of children with similar characteristics but who were not enrolled in the center. Methods. A quasi-experimental study was designed to observe 444 children aged 3 to 72 months from a low-income neighborhood in the city of Sorocaba, in the state of São Paulo, Brazil. Two groups were studied: 164 children enrolled in a local municipal day-care center (intervention group) and 280 not receiving care at the center (nonintervention, comparison group) but instead being cared for at home. Both groups were seen four times over a period of 16 months. At each observation session, the children's weight and height were measured. Information was also collected on the mother's sociodemographic characteristics and the illnesses she had suffered as well as the child's weight and other health characteristics at birth, the child's illnesses in the 15 days before each observation, and any hospitalizations. Results. The children in both groups were from low-income families, with 65% of the families having an average monthly income below US$ 100; 80% of the mothers had received 8 years of schooling or less. Multivariate linear regression analysis showed that at the first observation (just before enrollment in the day-care center), birth weight was the only factor that explained the nutritional differences between the two groups. Subsequent analyses showed that being in day care was the factor that best explained the differences between the groups, especially in terms of the adequacy of weight for age, after controlling for birthweight, sex, age at the beginning of the study, and illnesses in the 15 days before an observation session. The nutritional impact of the intervention was significant as early as 3 months after being enrolled in day care. Conclusions. The nutritional benefits of the care provided at the center outweighed the negative effects sometimes seen in such centers, such as the greater morbidity that children in day-care centers often experience in comparison to children receiving care at home.
Resumo:
Jerked beef is officially defined as salted, cured and dried beef. Water activity (Aw), moisture, ash and residual nitrite are the physicochemical parameters that define this product identity and quality standards. In this work, the behavior of these parameters during jerked beef processing was evaluated and a significant correlation among them was revealed. These results allowed the establishment of statistical equations that enable the estimation of all the physicochemical parameters from the results obtained in the measurement of just one of them.
Resumo:
This study evaluated using of Multicriteria Evaluation in a GIS, specifically by Weighted Linear Combination Method for generation of map of priority areas for forest restoration in the initial part of River Pardo Basin, SP, in order to water resources conservation. Aiming to define criteria and restrictions it was used Participatory Techniques, and the following factors had been selected: proximity of the hydrographic network, proximity of forest cover, slope and erodibility of soil. To calculate the weight to each factor it was used the decision-making process, known as Analytic Hierarchy Analysis, this method consists of a paired comparison of factors to determine the relative importance of each. According to Weighted Linear Combination, the very high priority areas have a more limited spatial distribution, with an apparent concentration around the water bodies, outlining a buffer to the river system. The proximity factor of the hydrographic network, and enables the connection forestry, contributed, along with the factor of proximity to forest cover, so there would be the definition of most of the areas with the highest priority in the basin, which concentrate the largest areas of forest and native riparian areas along the hydrographic.
Resumo:
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
Resumo:
Pós-graduação em Agronomia (Energia na Agricultura) - FCA
Resumo:
Pós-graduação em Agronomia (Energia na Agricultura) - FCA
Resumo:
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Resumo:
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)