Multivariate classification of growth models for lambs genetic groups

Abstract

The main objective of this work was to use the cluster analysis in order to classify nonlinear growth models in relation to different quality fit evaluators when data from the following lambs genetic groups were utilized: Dorper x Morada Nova, Dorper x Rabo Largo e Dorper x Santa Inês. After the choice of the best model, it was also aimed to apply the identity model test in order to identify the most efficient genetic group. The proposed methodology was applied to data of all animals from each group regarding twelve nonlinear models, whose fit quality was measured by determination coefficient, Akaike information criterion, Bayesian information criterion, mean quadratic error of prediction, predicted determination coefficient and convergence percentual. The cluster analysis indicated the von Bertalanffy as the best model for the three data sets. The model identity tests revealed that the Dorper x Santa Inês group presented higher adult weight, therefore this group is recommend for meat production.