Authors: Mehmet N. ORMAN, Fetih YILDIRIM
Abstract: In this study, the parameters of the regression models for the lactation curve were estimated. A total of 45 standard milk yield records obtained from the first, the second and the third lactations of Jersey bred in Karaköy State Farm were used in the present study. Milk production records were corrected for the ages of cows ond the numbers of lactation. The parameters were estimated in two stages. In the first stage, the data were applied to the five nonlinear models, developed by different authors. The results out of the five models were discussed in detail. The best fitting model is chosen according to the advantages and disadventages of these models. In the second stage, five different estimation methods were compared. Among the five methods studied, the MRT (Marquardt) method was seen to be more practical and required less computing time than the others. The best fit model was used with MRT method for the estimation of the parameters. The parameters, which related to the milk yield records, were estimated and the lactation curves were obtained for each of the theree lactation groups. The curves of the three lactation groups were discussed in regard to the animal breeding and statistics.
Keywords: Constrained Nonlinear Astimation, Least Squares, Lactation Curve
Full Text: PDF