Performance comparisons of some nonlinear functions in describing the growth curve of Zandi sheep breed

In this research fitness of nonlinear models of Von Bertalanffy, Gompertz, Brody and Logistic to describe the growth curve of Zandi sheep breed of Khojir station was studied. In this order from 14569 body weight records (from birth to 400 days of age) which have been recorded as daily from 3581 heads during the years 1992 to 2014 were used. Each model was fitted separately to body weight records using whole data, and also for different environmental factors (i.e., sex, type of birth, season of birth, year of birth and age of dam) separately using the nonlinear least square (NLIN) procedure of SAS. Goodness of fit of each model was determined using adjusted multiple coefficient of determination (R2Adj), Akaike’s information criterion (AIC), root mean square error (RMSE) and Durbin-Watson (DW). All used models in the current study fitted the growth data of Zandi sheep well based on different goodness of a fitting criteria. However, the results showed that Logistic growth model with the highest accuracy (R2Adj= 0.9702; AIC= 85886) and the lowest error (RMSE= 4.61) could describe growth curve better than the other growth models, and was followed by Gompertz, Von Bertalanffy and Brody growth models, respectively. The results of this research indicate that the Logistic model can be used to the regulate feeding programs, determination of management problems and optimum slaughtering age of lambs at the Khojir breeding situation.