Definition of subgroups for fixed regression in the test-day animal model for milk production of Holstein cattle in the Czech Republic
Zavadilova, L.(Vyzkumny Ustav Zivocisne Vyroby, Prague (Czech Republic))E-mail:dedkova.ludmila@vuzv.cz | Nemcova, E. | Pribyl, J. | Wolf, J.
The investigation was based on roughly 3.9, 2.7, and 1.7 million test-day records from first, second, and third lactation, respectively, sampled from 596,200 Czech Holstein cows between the years 1991 and 2002. Breeding values were estimated from multi-lactation random-regression test-day models which contained the fixed effect of herd-test day, fixed regression on days in milk, random regressions on the animal level, and the permanent environmental effect. Third degree Legendre polynomials were used for both the fixed and random regressions. The models differed in fixed regression. In Analysis I, 96 subclasses were defined according to age at calving, season, and year of calving within lactation. In Analysis II, days open were additionally included as a grouping factor resulting in 480 subclasses. Rank correlations over 0.98 between both analyses were observed for breeding values for sires. Grouping according to Analysis I was recommended.
显示更多 [+] 显示较少 [-]