Bayesian approach for the evaluation of adaptability and stability of alfalfa genotypes

• Autor(a) principal: Nascimento, Moysés
• Data de Publicação: 2011
• Outros Autores: Silva, Fabyano Fonseca, Sáfadi, Thelma, Nascimento, Ana Carolina Mota Campana, Ferreira, Reinaldo de Paula, Cruz, Cosme Damião
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Pesquisa Agropecuária Brasileira (Online)
• Texto Completo: https://seer.sct.embrapa.br/index.php/pab/article/view/8320
• topic: Medicago sativa; Bayes factor; informative prior; genotype x environment interaction; MCMC; Medicago sativa; fator de Bayes; priori informativa; interação genótipo x ambiente; MCMC

The objective of this work was to propose a Bayesian approach for the Eberhart & Russell method to evaluate the phenotypic adaptability and stability of alfafa (Medicago sativa) genotypes, as well as to evaluate the efficiency of the use of prior informative and noninformative distributions. Data from a randomized block design experiment evaluating the forage dry weight of 92 genotypes were used. The Bayesian methodology proposed was implemented in the free software R by the MCMCregress function of the MCMCpack package. To represent the noninformative prior distributions, a probability distribution with high variance was used; and, to represent the informative prior, a meta‑analysis concept was adopted, characterized by the use of information provided by previous studies. The comparison between the prior distributions was done using the Bayes Factor, with the BayesFactor function of the MCMCpack package, which indicated that an informative prior is more appropriate under the conditions of this study.