A longitudinal study of sweet orange flowering with grouped count data

• Autor(a) principal: Lara, Idemauro Antonio Rodrigues de
• Data de Publicação: 2020
• Outros Autores: Taconeli, Cesar Augusto, Moral, Rafael de Andrade, Hinde, John, Voigt, Vanessa, Freitas, Silvia Maria de
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Acta Scientiarum. Agronomy (Online)
• Texto Completo: http://www.periodicos.uem.br/ojs/index.php/ActaSciAgron/article/view/43734
• topic: citrus; discrete data; mixed model; Bayesian analysis; Probabilidade e Estatística Aplicadas; Citrus; discrete data; mixed model; Bayesian analysis.

The orange variety “x11”, which is a spontaneous mutant of the sweet orange, has a short juvenile period with early flowering. The data used in this paper are from a randomized design experiment that aimed to assess the plants' flowering characteristics when grafted onto two different varieties of lemon rootstock. The plants were pruned in each of the four seasons, and on each pruning occasion, the number of branches on each plant was counted and classified into four mutually exclusive flowering categories. The data presented large variability and many zeros. The statistical analysis included the use of generalized linear mixed models with a Bayesian approach. The results showed that flowering is not equal over the seasons, i.e., there are significant differences in the classification of the branches across the four seasons and the two varieties, with interactions between seasonal and branch effects.