dissertação
Estimação do tempo de latência em um modelo não-linear de degradabilidade in situ
Carregando...
Notas
Data
Autores
Orientadores
Editores
Coorientadores
Membros de banca
Título da Revista
ISSN da Revista
Título de Volume
Editor
UNIVERSIDADE FEDERAL DE LAVRAS
Faculdade, Instituto ou Escola
Departamento
Programa de Pós-Graduação
DEX - Departamento de Ciências Exatas
Agência de fomento
Tipo de impacto
Áreas Temáticas da Extenção
Objetivos de Desenvolvimento Sustentável
Dados abertos
Resumo
Savian, T.V. Estimação do tempo de latência em um modelo não-linear de degradabilidade em situ. 2005.55p. Dissertação(Mestrado em Estatística Agropecuária)-Universidade Federal de Lavras, Lavras, 2005.
We proposed evaluating the parameters of degradation model developed by Mertens & Loften (1980) adjusted for data come from a trial of in situ degradability. The experiment has evaluated the coast-cross´s (Cynodon dactylon x Cynodon nlemfunensis) potentially degradable residues of neutral detergent fiber (NDF) on three cutting ages (30, 60, 90 days), with three replications. At each cutting age, NDF potentially degraded residues was investigated by utilizing fifteen incubation times (0; 0.5; 3, 6, 12, 24, 48, 72, 96 and 120 hours). Non-lactating cows with permanent rumen cannula constituted the experimental unit. The animal´s individual and mean adjustments were obtained by using the following models: weighted by the inverse of variance without autoregressive error (AR), non-weighted with AR and non-weighted without AR. It was estimated the variances of the parameters estimators by covariance matrix and by jackknife method. It was calculated expressions for the confidence interval for the model parameters. The results showed that the model weighted by the inverse of variance provided estimates equally to zero for the lag time. The consideration of second order autoregressive error improved the fit of Mertens & Loften (1980) model, promoting more accurate estimates of the parameters. The jackknife method showed larger variance estimates, for all cutting ages, both individual and mean adjustments.
We proposed evaluating the parameters of degradation model developed by Mertens & Loften (1980) adjusted for data come from a trial of in situ degradability. The experiment has evaluated the coast-cross´s (Cynodon dactylon x Cynodon nlemfunensis) potentially degradable residues of neutral detergent fiber (NDF) on three cutting ages (30, 60, 90 days), with three replications. At each cutting age, NDF potentially degraded residues was investigated by utilizing fifteen incubation times (0; 0.5; 3, 6, 12, 24, 48, 72, 96 and 120 hours). Non-lactating cows with permanent rumen cannula constituted the experimental unit. The animal´s individual and mean adjustments were obtained by using the following models: weighted by the inverse of variance without autoregressive error (AR), non-weighted with AR and non-weighted without AR. It was estimated the variances of the parameters estimators by covariance matrix and by jackknife method. It was calculated expressions for the confidence interval for the model parameters. The results showed that the model weighted by the inverse of variance provided estimates equally to zero for the lag time. The consideration of second order autoregressive error improved the fit of Mertens & Loften (1980) model, promoting more accurate estimates of the parameters. The jackknife method showed larger variance estimates, for all cutting ages, both individual and mean adjustments.
Abstract
Descrição
Área de concentração
Estatística e Experimentação Agropecuária
Agência de desenvolvimento
Palavra chave
Marca
Objetivo
Procedência
Impacto da pesquisa
Resumen
ISBN
DOI
Citação
Savian, T.V. Estimação do tempo de latência em um modelo não-linear de degradabilidade em situ. 2005.55p. Dissertação(Mestrado em Estatística Agropecuária)-Universidade Federal de Lavras, Lavras, 2005.