Seleção e classificação multivariada de modelos de crescimento não linearespara bovinos Nelore

Abstract

This study aimed to evaluate cluster analysis in classifying and selecting non linear models to describe Nelore beef cattle growth based on different goodness of fit criteria tests. A total of 12 non linear models were evaluated based on the following criteria: the determination coefficient (R2), error mean square (QME), Akaike information criterion (AIC), Bayesian information criterion (BIC), mean quadratic error of prediction (MEP) and predicted determination coefficient (R2p). The Brody model showed the best adjustment for the data set.