Regressão não linear multivariada no crescimento do coco variedade anã verde

• Ano de defesa: 2016
• Autor(a) principal: Prado, Thalita Kelen Leal do
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Lavras; Programa de Pós-Graduação em Estatística e Experimentação Agropecuária,; UFLA; brasil; Departamento de Ciências Exatas
• Palavras-chave em Português: Modelo de crescimento; Análise de regressão multivariada; Mínimos quadrados; Growth model; Multivariate regression analysis; Least squares; Probabilidade e Estatística Aplicadas
• Link de acesso: http://repositorio.ufla.br/jspui/handle/1/10872

The green dwarf coconut palm is a major plant resource of humanity whose operation has evolved in most brazilian states because the commercial interest in coconut juice for fresh consumption and use in potting industry, taking up space on bulky soft drinks market. In commercial plantations for the water market in Brazil dominates the dwarf variety, because of its good performance in terms of yield and quality of coconut water. The nonlinear regression models logistic and gompertz growth in the description of plants and fruits have been used. The objective of this study was to evaluate the fit of nonlinear logistic multivariate models (LL), gompertz (GG) and hybrid (GL and LG) by the method of least squares, structure basis of independent and autoregressive errors of the second order for the residuals, to obtain estimates of the parameters. The multivariate models were adjusted growth data of green dwarf coconut fruit, longitudinal and transverse outer diameter. Choosing the best model was made using the Akaike information criterion corrected coefficient of determination adjusted and residual mean square and all the models showed a good fit and the multivariate model gompertz GG, considering auto regressive structure of the second order, were more adequate to fit the experimental data, resulting in estimates consistent with those reported in the literature. The adjustment procedures regression models were performed by the computer program Statistical Analysis System SAS.