Comparação entre métodos univariados e multivariados na seleção de variáveis independentes, na construção de tabelas volumétricas para Leucaena leicocephala (Lam) de Wit

• Ano de defesa: 2005
• Autor(a) principal: ARAÚJO, Adalberto Gomes de
• Orientador(a): SILVA, José Antônio Aleixo da
• Banca de defesa: MARINO, Jacira Guiro, MENDES, Paulo de Paula
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal Rural de Pernambuco
• Programa de Pós-Graduação: Programa de Pós-Graduação em Biometria e Estatística Aplicada
• Departamento: Departamento de Estatística e Informática
• País: Brasil
• Palavras-chave em Português: Análise univariada; Análise multivariada; Leucaena leucocephala
• Palavras-chave em Inglês: Univariate analysis; Multivariate analysis
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::PROBABILIDADE E ESTATISTICA
• Link de acesso: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/4450

The objective of this work was to use multivariate and univariate statistical methods, in the selection of independent variables, in mathematical models, in the construction of volume tables for Leucaena leucocephala, looking for reduction in time and costs, without loss of precision. The data came from an experiment carried out at the Experimental Station of the Institute of Agriculture Research (IPA), Caruaru-PE. It was used 201 trees of leucena that had their volumes (dependent variable) measured by the method of Smalian, and 20 variables independent measured in the same trees. For the selection of the independent variables the following methods were used: Principal Components, Cluster Analysis, Maximum and Minimum R2, Stepwise, Forward, Backward and Criterion of Akaike. In the general, the univariate and multivariate methods used in the selection of independent variables for volume models, showed similar responses, even though they had different structures in relation to the independent variables, since the number of those variables is high. Besides the applied statistical tests, the researcher'sjudgment about the relevance of the selected independent variables in the final equations has a great importance, mainly, in the reduction of costs and sampling errors.