Modelos de regressão com efeitos mistos aplicados a dados de germinação de sementes

Detalhes bibliográficos
Ano de defesa: 2017
Autor(a) principal: Pereira, Edilenia Queiroz
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Tese
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Estadual de Maringá
Brasil
Departamento de Estatística
Programa de Pós-Graduação em Bioestatística
UEM
Maringá, PR
Centro de Ciências Exatas
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: http://repositorio.uem.br:8080/jspui/handle/1/4366
Resumo: Stevia rebaudiana is a plant that has great potential when it comes to the production of natural sweeteners, this fact produces a great economic and scientific interest in relation to its characteristics and therapeutic properties, which are present in great majority in its sheets. Interested in understanding the germination behavior of Stevia rebaudiana seeds under controlled conditions, researchers from the Experimental Farm of Iguatemi conducted an experiment with 9,600 seeds taken from four different lots and submitted to light and position effects Within the germinator us. In order to analyze these data, we used the generalized linear mixed regression models with Binomial distribution, Beta-binomial with frequentist approach and Bayesian approach and the Multinomial distribution with frequentist approach. The estimation and validation of the adjusted models were performed by the frequentist package "gamlss" and by Bayesian package "INLA" of the statistical environment R. The logit link function was used in all fitted models and the choice of the Final model was given by means of AIC, BIC and Deviance Global measurements from the frequentist point of view and DIC, WAIC and LPML from the Bayesian point of view. By the criteria of selection of models it was noticed that the Binomial and Beta-binomial models did not present difference in relation to the quality of the obtained results, being thus equivalent, both by frequentist point of view as Bayesian. The methodology used was adequate in explaining the germination, in relation to the factors of influence or explanatory variables, namely light, position, lot and germinator. Based on the adjusted models it was possible to identify that there is no difference in seed germination among the four germinators used. On the other hand, both the type of light applied and the position in which these seeds are placed inside the germinators have been influencing the germination results of the seeds.