Comparison of non-linear models to describe growth of Iranian Guilan sheep

Summary Background: non-linear mathematical models, empirically developed by plotting body weight against age, are used to describe the growth curve in different animals. Objective: to describe the growth pattern in Guilan sheep using non-linear models. Methods: six non-linear mathematical equations (Brody, Negative exponential, Logistic, Gompertz, von Bertalanffy and Richards) were used to describe the growth curves in Guilan sheep. The Agricultural Organization of Guilan province (Rasht, Iran) provided the dataset used in this study. The dataset included 42,257 weight records of lambs from birth to 240 days of age during years 1994 to 2014. Each model was separately fitted to body weight records of all lambs, males and females, using the NLIN and MODEL procedures of SAS. The models were tested for goodness of fit using adjusted coefficient of determination, root means square error (RMSE), Durbin-Watson statistic, akaike’s information criterion (AIC) and bayesian information criterion (BIC). Results: the Richards model provided the best fit to the growth curve in females and all lambs, with the lowest RMSE, AIC, and BIC values compared to the other models. The Brody model provided the best fit of growth in male lambs due to the lower values of AIC and BIC compared to the other models. The negative exponential model provided the worst fit of growth for males, females and all lambs. Conclusion: the evaluation of the growth equations used in this study indicates the potential of non-linear functions for fitting body weight records of Guilan sheep