Modelo Logístico Bayesiano no estudo do crescimento de tomates

Abstract

Knowing the growth of tomato and its fruits, as measured by biomass accumulation over time is essential for the proper handling and detection of problems in the development of crops. This growth can be studied using various models of non-linear regression that can be used to facilitate interpretation of the processes involved in plant production system. Among the empirical models often used to estimate the growth of plants and their components is the function Logistic. One method used to estimate the parameters of the growth rate is the Bayesian method. The study objective to apply the Bayesian approach in describing the data – real and simulated – the diameter growth of tomatoes, using the model Logistic. The algorithms for the Gibbs Sampler and Metropolis – Hastings were implemented using the R language. The condition of convergence of the chains was checked using the criteria suggested by Nogueira, Sáfadi and Ferreira (2004) available on the R software package BOA. The Bayesian approach was efficient, since it was evaluated and verified by the simulation process, with very close estimates of the parametric value, and estimates were shown to be consistent with the values reported in literature.