Eberhart and Russel's Bayesian Method in the Selection of Popcorn Cultivars

The goal of this work was to estimate stability and adaptability parameters using a Bayesian approach to Eberhart and russel's method and to assess the efficiency of using an a priori dis - tribution. The information from assessing the popping expansion and grain yield of 16 pop - corn genotypes was used in randomized block experiments implemented in five environments in the North and Northeast regions of the State of rio de Janeiro, Brazil. The Bayesian method - ology was implemented using the free software package r with the MCMCregress function of the MCMCpack package. Eberhart and rus- sel's method using a Bayesian technique was found to be efficient in recommending cultivars to more or less favorable environments. The incorporation of a priori information provided greater accuracy in estimating the stability and adaptability parameters. In the comparison of a priori distributions, the BayesFactor function indicated the informative a priori as the most effective for obtaining reliable estimates.