Abstract
A study has been conducted to compare the three growth curve models for fitting to growth data in wild and laboratory mice. Body weight data from birth to ten weeks of age were taken weekly from two subspecies of mice, Yonakuni wild mouse (Yk, Mus musculus molossinus yonakuni) and laboratory outbred mouse of CF#1 (Mus musculus domesticus) . The numbers of Yk were 44 males and 45 females, whereas of CF#1 were 45 males and 45 females. Three growth curve models, namely Gompertz, logistic and Asymptotic models, were used to describe growth patterns within each subspecies-sex subclass. All models were fitted to weight data by Gauss-Newton method of SAS. A criterion based on coefficient of determination (R2) has been used for comparing the three growth curve models. High value of R2 indicates goodness of fit attained by the model. Mean R2 values in fitting Gompertz model were 0.999 for both subspecies at two sexes, indicating that this model was the best fit to body weight data in mice used.
