Evaluation of the growth curve by the method of iterated least square
2 Ankara Üniversitesi, Veteriner Fakültesi, Biyometri Anabilim Dalı, Ankara - This study was conducted to find the best fitted stastical model for the growth poultry.
The biological material of the study consisted of Broiler breeding female chicks which hatched in December, May and October. The growth of chickens was measured by weekly live weigth until 63 th months of age. The data was processed using six non-linear growth models. The statgraph program was used for the statistical procedure.
The results revealed that the chicks hatched in May had nearest values to the combined material.
The results showed that Richards growth model was the best model in explaining the growth of the chicks for the three hatching periods and also for the combined material.
The models of Brody, Bertalanffy, Stevens, Gompertz and Logistic followed Richards model in the descending order. The fitting criteria were a value, error sum of squares (HKT) and determination coefficient. In the Richards model the values for , HKT an r2 were 3141.12, 63746.65 and 0.9972, respectively. The error sum of squares for the other models were 65494.7, 65496.6, 127281.9, 192664.8, 423222.9, in the above order.
Keywords :