The study was performed to model growth curves of crossbred dairy cattle using different non-linear models. All available weight-age records of crossbred dairy cattle, collected from birth to 1008 days of ages during the thirty-two years periods (1985-2016) were used for the study. Nonlinear models were subjected to fit the data using the NLIN procedure of SAS ver. 9.40 (2013). The model(s) that best fit and describe the curve characteristics were chosen based on goodness of fit statistics. The study revealed that the asymptotic mature body weight (A) varied from 413.90±7.48 kg for Logistics to 1087.80±125 kg for Von Bertalanffy models. The highest maturation rate (0.0059±0.0002) was found in the Logistics model while the smallest maturation rate was observed in Quadratic (-0.0002±4×10-4) models. The values of asymptotic mature weight (B) ranged from 0.3096±0.01 to 9.766±0.55 for Cubic and Logistics models, respectively. The Quadratic model gave highest fit to the weight-age data with R2 value of 1.0. The Gompertz, Cubic and Logistics models had also better fitted the observed weight-age data with the highest R2 values. Quadratic, Gompertz, Cubic and Logistics models are recommended for estimating later live body weight of the cattle from birth to 1008 days of age.

Keywords: Growth curve, Non-linear model, Modeling, Crossbred dairy cows