1000 resultados para INFERÊNCIA ESTATÍSTICA
Resumo:
Foram avaliadas três formas de parcelas experimentais (retangular, uma linha - linear e parcela de uma árvore - STP) em testes clonais de Eucalyptus spp, utilizando-se três experimentos, cada um com 18 clones. Foram usados três modelos de análise (mínimos quadrados ordinários - ANOVA tradicional, modelos mistos com fator clone fixo ou com fator clone aleatório - REML/BLUP). Os dois primeiros modelos apresentaram resultados similares. Com REML/BLUP houve estreitamento das predições em relação às amplitudes obtidas com as médias, e essa redução foi proporcionalmente maior com parcelas retangulares e STP. O ordenamento dos clones também foi similar com esses dois tipos de parcelas. É provável que com parcelas STP haja um balanço compensatório das alocompetições, pois se pode trabalhar com maior número de repetições e menor custo. Portanto, com parcelas STP haverá economia de recursos e sem prejuízos para o Programa de Melhoramento Florestal.
Resumo:
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Resumo:
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
Resumo:
Objetivou-se estimar parâmetros genéticos, utilizando inferência Bayesiana, para as estimativas dos parâmetros individuais de peso à maturidade (Â) e taxa de crescimento, obtidos pela função de crescimento Brody. O arquivo estava constituído de 14.563 registros de pesos e idades referentes a 1.158 fêmeas da raça Nelore, participantes do Programa de Melhoramento Genético da Raça Nelore. Para a análise das estimativas dos parâmetros da curva, via inferência bayesiana, foi proposto um modelo animal unicaráter, que incluiu como fixo o efeito de grupo contemporâneo (animais nascidos no mesmo estado, no mesmo trimestre do ano, mesmo ano e mesmo regime alimentar) e como aleatórios os efeitos genético direto e residual. Nessa análise, foram utilizados dois diferentes tamanhos para as cadeias geradas pelo algoritmo de amostragem de Gibbs, de 550 e 1.100 mil ciclos, com períodos de descarte amostral de 50 e 100 mil ciclos, respectivamente, e amostragens a cada 500 e 1.000 ciclos, respectivamente. As médias posteriores da variância genética aditiva e residual foram próximas, tanto para  quanto para a, mesmo quando implementados diferentes tamanhos para as cadeias geradas pelo algoritmo de amostragem de Gibbs. Os coeficientes de herdabilidade estimados para Â, variaram de 0,44 a 0,46, amplitude semelhante aos 0,46 a 0,48 obtidos para as estimativas de. Essas magnitudes indicam que a seleção pode ser usada como instrumento para alterar a forma da curva de crescimento desses animais. Entretanto, o uso das informações obtidas, visando à alteração da curva de crescimento dos animais, deve ser feito com grande cautela, uma vez que as características a serem trabalhadas na modificação do formato da curva de crescimento, de acordo com resultados da literatura especializada, são negativamente correlacionadas.
Resumo:
Apresenta-se um sistema computacional, denominado ICADPLUS, desenvolvido para elaboração de banco de dados, tabulação de dados, cálculo do índice CPO e análise estatística para estimação de intervalos de confiança e comparação de resultados de duas populações.Tem como objetivo apresentar método simplificado para atender necessidades de serviços de saúde na área de odontologia processando fichas utilizadas por cirurgiões dentistas em levantamentos epidemiológicos de cárie dentária. A característica principal do sistema é a dispensa de profissional especializado na área de odontologia e computação, exigindo o conhecimento mínimo de digitação por parte do usuário, pois apresenta menus simples e claros como também relatórios padronizados, sem possibilidade de erro. Possui opções para fichas de CPO segundo Klein e Palmer, CPO proposto pela OMS, CPOS segundo Klein, Palmer e Knutson, e ceo. A validação do sistema foi feita por comparação com outros métodos, permitindo recomendar sua adoção.
Resumo:
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
Resumo:
The objective of this study was to estimate genetic parameters for body weights at weaning (PD), 12 months old (P12) and adult age (PAD), culling age (TPR, days in herd), number (ND10) and kilograms (QD10) of calves weaned up to ten years of age, total number (NDT) and total kilograms (QDT) of calves weaned during herd life, and kilograms of calves weaned per year in herd (QTPR) of Canchim (5/8 Charolais + 3/8 Zebu) females from one herd. Data consisted of 3,249, 3.111, 1,138, 1,340, 1,362, 1,362, 1,340, 1,340 and 1,340 records of PD, P12, PAD, TPR, ND10, QD10, NDT, QDT and QTPR. respectively. Variance and covariance components were estimated by bivariate analyses between PD, P12 and PAD and other production traits using Bayesian inference. The models included the additive direct, permanent environmental and residual random effects and the fixed effects year and month of birth or calving, calving age and age of the animal, depending on the trait. QD10, QDT and QTPR of each female were obtained by adjusting the weaning weights of calves for year and month of birth, sex and age of cow. Average of heritability estimates were 0.38 (PD), 0.40 (P12), 0.54 (PAD), 0.22 (TPR), 0.22 (ND10), 0.24 (QD10), 0.23 (NDT), 0.23 (QDT) and 0.32 (QTPR), indicating genetic variability to obtain response by selection. Genetic correlations between TPR (-0.02, 0.26 and -0.12), ND10 (0.04, 0.10 and -0.29), QD10 (0.37, 0.39 and -0.13), NDT (-0.03, 0.14 and -0.25), QDT (0.20, 0.33 and -0.16), QTPR (0.21, 0.28 and -0.19) and body weights (PD, P12 and PAD) suggest that selection of females based on weaning and 12-month body weights will not affect productivity. However, it may be decreased by increasing female adult body weight.
Resumo:
The objective of this study was to evaluate the genotype x environment interaction for weaning and yearling weights, daily weight gain from weaning to 12 months of age and the growth performance in Canchim (5/8 Charolais + 3/8 Zebu) beef cattle estimated by a principal components analysis including those three traits. The environment was defined by season of birth (first and second semesters of the year). Genetic parameters were estimated by bayesian method with the Gibbs sampler using bivariate analyses (considering the trait in each of the two seasons as a different one) and models that included the fixed effects of year and month of birth, sex and age of cow (linear and quadratic) and the random effects of animal and residual. The results suggested that genetic evaluation and selection in Canchim beef cattle for the traits studied should consider the genotype and season of birth interaction.
Resumo:
There are several papers on pruning methods in the artificial neural networks area. However, with rare exceptions, none of them presents an appropriate statistical evaluation of such methods. In this article, we proved statistically the ability of some methods to reduce the number of neurons of the hidden layer of a multilayer perceptron neural network (MLP), and to maintain the same landing of classification error of the initial net. They are evaluated seven pruning methods. The experimental investigation was accomplished on five groups of generated data and in two groups of real data. Three variables were accompanied in the study: apparent classification error rate in the test group (REA); number of hidden neurons, obtained after the application of the pruning method; and number of training/retraining epochs, to evaluate the computational effort. The non-parametric Friedman's test was used to do the statistical analysis.
Resumo:
Data from purebred Simmental, Nellore and Canchim cattle breeds obtained from the respective Brazilian Associations of Breeders were used to estimate variance components and to predict genetic values for 365 days weight. The results obtained by Bayesian inference were compared to those from Restricted Maximum Likelihood (REML) and Best Linear Unbiased Prediction (BLUP), which are the most commonly used methods of estimation and prediction in animal breeding. The two methods presented similar point estimates but the study of the marginal posterior distributions in the Bayesian approach yields more detailed information about the parameters and other unknowns in the model.
Resumo:
The epidemiological surveys are important to the deployment, implementation and evaluation of projects and health actions in a community. The planning, goals, samples, team training/calibration, carrying out and publication of results are extremely important in the epidemiological surveys. Thus, the care with the sample and statistical analysis is fundamental for the results to be consistent and trustworthy in order to be able to be inferred for all the population. The aims of this study is to investigate the statistical methodology used in papers on dental caries epidemiological surveys published from 1960 to 2001. A bibliographical survey was carried out in BBO, MEDLINE and SCIELO databases. The papers found were analyzed with regards to the statistical methodology applied in the whole study, from the sampling to the tabling of data. Most studies (72.6%) only presented the number of elements that composed the sample, without explaining the planning involved in obtaining it.
Resumo:
This article deals with the development of three competencies defined and studied in the context of Statistics Education (literacy, thinking and statistical reasoning). These competencies based on interpretation and understanding of critical information from real data are associated with an education geared toward the formation of critical citizenship and are in agreement with the principles that guide Critical Education. In this context, we consider Mathematical Modeling as a concrete possibility for integration between Critical Education and Education Statistics. We show that work with modeling projects in the classroom takes place in a context in which teaching is guided by the fundamentals of Critical Education. We present a project entitled Teaching Statistics and the financial market which illustrates the congruence between the objectives of Education Statistics and Critical Education.
Resumo:
This article presents an exercise in meta-comprehension of what has been researched on teaching probability and statistics in Brazil. This research was based on the work on this subject presented in the third International Symposium for Research in Mathematics Education (III SIPEM). Articles were selected from the proceedings of the event analyzed hermeneuticly according to the procedures of phenomenology. We observed no evidence of clustering of research on this topic in terms of region or institutions, and we also emphasize that research on the teaching of Probability and Statistics needs to advance toward a theoretical discussion that transcends the subjects being studied and makes broader and deeper links between theory and practice. Findings also indicate that this sub-area of research in mathematics education is in the process of constituting itself.
Resumo:
Incluye Bibliografía
Resumo:
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)