Methodology for the Evaluation of Crossbred Dairy Cattle. Comparison of the effectiveness of different models and methods for sire evaluation

      Аннотация: Efforts are being made at different places in India to record the performance of dairy animals. In view of the heterogeneous managemental practices among households under field conditions, the methodology developed and practiced for analysis of data under different conditions in other countries may not be exactly suitable for data under field conditions in India. Therefore, it is of primary importance to identify appropriate analytical models and methods to analyze the data on dairy animals managed under field conditions, so that genetically superior animals could be selected as breeding animals. The high product-moment and rank correlations between true values and the breeding values estimated by LS, SRLS, BLUP, and REML methods for different sizes of data sets at all the three levels of heritability indicated that these four methods are more or less similar in their accuracy for estimating the breeding values and ranking of bulls.

Ключевые слова: Heritability, simple daughters’ average (SD), contemporary comparison (CC), least squares (LS), simple regressed least squares (SRLS), best linear unbiased prediction (BLUP), restricted maximum likelihood (REML)