Comparison of Nonlinear Models to Describe the Growth of Tuj and Romanov x Tuj (F1) Lambs

Abstract

This study aimed to identify the most suitable model for explaining weight changes in purebred Tuj (n=35) and Romanov x Tuj (RoxTuj) (F1) (n=25) lambs using non-linear models. Single-born lambs of both breeds and genders were included in the evaluation. Five different non-linear growth models were compared: Brody, Gompertz, Logistic, Richards, and Weibull. The best model for describing growth was chosen based on four criteria: coefficient of determination (R²), mean square error (MSE), Akaike information criterion (AIC), and Bayesian information criterion (BIC). Models with the highest R² and the lowest MSE, AIC, and BIC values were considered the best fit for the data. It was observed that the Brody model had the highest R2 and lowest MSE, AIC and BIC values for Tuj and RoxTuj (F1) female and male lambs. The Gompertz, Logistic, and Richards models exhibited similar predictive performance. In contrast, the Weibull model produced significantly different results compared to the other models when predicting weight changes. Therefore, the Brody model was identified as the most effective model for explaining growth patterns in both Tuj and RoxTuj (F1) lambs.