GROWTH DESCRIPTION OF PURE AND CROSSBRED TURKEYS USING NON-LINEAR MODELS IN HOT AND HUMID TROPICAL ENVIRONMENT

This study was conducted to compare three non-linear growth models (Logistics, Gompertz and Von Bertalanffy) in describing the of 360 turkeys genotypes (exotic, crossbred local). The curve parameters - A, B K for each model were estimated turkey using NLIN procedure (Marquart algorithm) Bayesian approach with a fixed-effect model. best fit also estimated. A comparison done among Akaike Information Criteria (AIC) (BIC). Pearson product-moment correlation coefficient (r), coefficient, used examine linear relationship between two quantities, parameters. For parameter predicted, Bertalanffy's predicted smallest value both criteria all male local female turkeys, while Gompertz's had exotic turkeys. Based on goodness Model Choice (Deviance Criteria), showed fit. However, difference is very minimal, thus, making suitable modeling domestic consider how this weight attained.