Modelos não lineares mistos em estudos de crescimento de frango de corte

• Ano de defesa: 2018
• Autor(a) principal: Yamanaka, Sandra Emiko
• Orientador(a): Venezuela, Antonio Luís
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus Sorocaba
• Programa de Pós-Graduação: Programa de Mestrado Profissional em Matemática em Rede Nacional - PROFMAT
• Palavras-chave em Português: Modelo não Linear Misto; Medidas Repetidas; Variabilidade; Premixes
• Palavras-chave em Inglês: Mixed nonlinear model; Repeated measures; Variability; Premixes
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::MATEMATICA; CIENCIAS EXATAS E DA TERRA::MATEMATICA::MATEMATICA APLICADA
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/10380

Premix is the main source of micronutrients such as vitamins, probiotics, prebiotics, enzymes, and macro and microminerals in poultry feed, that are included in the ration to improv the digestion of food and the performance of the selected lineage. Therefore, the knowledge of these compounds has great importance in studies on the growth of broilers. Nonlinear models are widely used in these studies, seeking to estimate parameters related to adult weight and growth rate through of methods that consider correlation between the repeated measures in the same chicken and the heterogeneity of the variances of the responses. As individual differences can occur for the weight response of birds, the use of mixed nonlinear models is proposed, allowing the nonlinear function on fixed and random effects, which may solve the problem of explaining variation among these kind of birds. In this work two common alternatives of growth data analysis were used and their results were compared with the approach using the mixed nonlinear model. The non-linear Gompertz model was used to explain the relationship between body weight and age, in broilers fed with feed supplemented with three different brands of premixes available in the market. The measurements of the weighings were collected in seven rounds at baseline and at 1st, 2nd, 3rd, 4th, 5th and 6th weeks. There data were obtained from a completely randomized design experiment with three hundred and seventy chickens from the Cobb line, three treatments and seven replicates.