Ajuste do modelo logístico na descrição do crescimento de frutos de coqueiro anão por meio de algorítmos iterativos MCMC

Abstract

Logistic models are widely used to predict outcomes of growth based on one or more predicted values. To this end, a logistic model with autoregressive error was applied to study the growth of green dwarf coconut by sampling the likelihood function via Markov Chain Monte Carlo (MCMC), based on fruit outside longitudinal diameter (DEL) in different ages. Besides, a simulation study was conducted in order to assure the developed methodology with efficient results. The estimates were in accordance to the values reported in scientific literature for this variety of coconut. Results indicate the logistic model was adequate to describe the DEL growth, and could be inferred that the green dwarf coconut is totally developed when outside diameter ranges from 19.30 cm to 20.70 cm, with credibility of 95%.